Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks432 from public data
- 1,512
- Learning Fine-Grained Image Similarity with Deep Ranking1,376
A deep ranking model that employs deep learning techniques to learn similarity metric directly from images has higher learning capability than models based on hand-crafted features and deep classification models.
- 885
- Interpretable Convolutional Neural Networks855
A method to modify a traditional convolutional neural network into an interpretable CNN, in order to clarify knowledge representations in high conv-layers of the CNN, which can help people understand the logic inside a CNN.
- Vision-Based Gesture Recognition: A Review668
A survey on recent vision-based gesture recognition approaches is given and methods of static hand posture and temporal gesture recognition are reviewed, along with some thoughts about future research directions.
- Cross-View Action Modeling, Learning, and Recognition659
A novel multiview spatio-temporal and-or graph (MST-AOG) representation for cross-view action recognition, which takes advantage of the 3D human skeleton data obtained from Kinect cameras to avoid annotating enormous multi-view video frames, but the recognition does not need 3D information and is based on 2D video input.
- 531
- Minimax Entropy Principle and Its Application to Texture Modeling522
The minimax entropy principle is applied to texture modeling, where a novel Markov random field model, called FRAME, is derived, and encouraging results are obtained in experiments on a variety of texture images.
- Robust 3D Action Recognition with Random Occupancy Patterns501
This work extracts semi-local features called random occupancy pattern ROP features, which employ a novel sampling scheme that effectively explores an extremely large sampling space and utilizes a sparse coding approach to robustly encode these features.
- Multi-Agent Tensor Fusion for Contextual Trajectory Prediction477
This work encodes multiple agents' past trajectories and the scene context into a Multi-Agent Tensor, then applies convolutional fusion to capture multiagent interactions while retaining the spatial structure of agents and thescene context.
- Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoning476
A novel approach is proposed, PromptPG, which utilizes policy gradient to learn to select in-context examples from a small amount of training data and then constructs the corresponding prompt for the test example, which verifies its effectiveness in selecting in- context examples.
- 476
- Interpreting CNNs via Decision Trees356
The proposed decision tree is a decision tree, which clarifies the specific reason for each prediction made by the CNN at the semantic level, and organizes all potential decision modes in a coarse-to-fine manner to explain CNN predictions at different fine-grained levels.
- 283
- 263
- Interpreting CNN Knowledge via an Explanatory Graph255
This paper learns a graphical model, namely an explanatory graph, which reveals the knowledge hierarchy hidden inside a pre-trained CNN in an unsupervised manner, i.e., without a need of annotating object parts.
- 254
- Deeply Learned Compositional Models for Human Pose Estimation253
A novel framework, termed as Deeply Learned Compositional Model (DLCM), is introduced, which exploits deep neural networks to learn the compositionality of human bodies and proposes a novel bone-based part representation that not only compactly encodes orientations, scales and shapes of parts, but also avoids their potentially large state spaces.
- 240
- Uncertainty-Aware Score Distribution Learning for Action Quality Assessment202
An uncertainty-aware score distribution learning (USDL) approach for action quality assessment (AQA) is proposed, which regards an action as an instance associated with a score distribution, which describes the probability of different evaluated scores.
- 200
- On the Anatomy of MCMC-Based Maximum Likelihood Learning of Energy-Based Models195
This study investigates the effects of Markov chain Monte Carlo (MCMC) sampling in unsupervised Maximum Likelihood (ML) learning and identifies a variety of ML learning outcomes that depend solely on the implementation of MCMC sampling.
- 188
- Learning Active Basis Model for Object Detection and Recognition183
An active basis model, a shared sketch algorithm, and a computational architecture of sum-max maps for representing, learning, and recognizing deformable templates are proposed.
- 178
- Dark, Beyond Deep: A Paradigm Shift to Cognitive AI with Humanlike Common Sense169
It is argued that the next generation of AI must embrace "dark" humanlike common sense for solving novel tasks, and is demonstrated how to observe and apply FPICU with little training data to solve a wide range of challenging tasks, including tool use, planning, utility inference, and social learning.
- 166
- A Modulation Module for Multi-task Learning with Applications in Image Retrieval157
A general modulation module is proposed, which can be inserted into any convolutional neural network architecture, to encourage the coupling and feature sharing of relevant tasks while disentangling the learning of irrelevant tasks with minor parameters addition.
- Learning Descriptor Networks for 3D Shape Synthesis and Analysis157
A 3D shape descriptor network, which is a deep convolutional energy-based model, for modeling volumetric shape patterns and can be used to train a 3D generator network via MCMC teaching.
- 156
- 153
- A tale of two explanations: Enhancing human trust by explaining robot behavior152
A psychological experiment to examine what forms of explanations best foster human trust in the robot found that comprehensive and real-time visualizations of the robot’s internal decisions were more effective in promoting human trust than explanations based on summary text descriptions.
- Cooperative Training of Descriptor and Generator Networks152
It is shown that the cooperative algorithm can learn highly realistic generative models, including a deep energy-based model and a non-linear version of factor analysis.
- Robust Visual Tracking by Integrating Multiple Cues Based on Co-Inference Learning151
This paper formulates the problem of multiple cue integration and tracking in a probabilistic framework based on a factorized graphical model and proposes a sequential Monte Carlo algorithm to provide an efficient simulation and approximation of the co-inferencing of multiple cues.
- 149
- 146
- Visual panel: virtual mouse, keyboard and 3D controller with an ordinary piece of paper140
By detecting the clicking and dragging actions, the system can fulfill many tasks such as controlling a remote large display, and simulating a physical keyboard, as well as providing 3D information, serving as a virtual joystick, to control 3D virtual objects.
- 139
- Exploring texture ensembles by efficient Markov chain Monte Carlo-Toward a "trichromacy" theory of texture136
The intrinsic link between the scientific definition of texture on Z/sup 2/ and the mathematical models of texture on finite lattices is established and an efficient Markov chain Monte Carte algorithm is proposed for sampling Julesz ensembles.
- 134
- Flow Contrastive Estimation of Energy-Based Models132
A significant improvement on the synthesis quality of the flow model is demonstrated, and the effectiveness of unsupervised feature learning by the learned energy-based model is shown, which can be easily adapted to semi-supervised learning.
- LLM 3 : Large Language Model-based Task and Motion Planning with Motion Failure Reasoning131
LLM3, a novel Large Language Model (LLM)-based TAMP framework featuring a domain-independent interface, leverages the powerful reasoning and planning capabilities of pre-trained LLMs to propose symbolic action sequences and select continuous action parameters for motion planning.
- 130
- 126
- Trajectory Prediction with Latent Belief Energy-Based Model121
The LB-EBM model is able to make accurate, multi-modal, and social compliant trajectory predictions and improves over prior state-of-the-arts performance on the Stanford Drone trajectory prediction benchmark by 10.9% and on the ETH-UCY benchmark by 27.6%.
- 121
- Primal sketch: Integrating structure and texture119
A generative image model, which is called ''primal sketch,'' following Marr's insight and terminology is proposed, which combines two prominent classes of generative models, namely, sparse coding model and Markov random field model for representing geometric structures and stochastic textures, respectively.
- 113
- 111
- Latent Diffusion Energy-Based Model for Interpretable Text Modeling107
A novel symbiosis between the diffusion models and latent space EBMs in a variational learning framework is introduced, coined as the latent diffusion energy-based model, which develops a geometric clustering-based regularization jointly with the information bottleneck to further improve the quality of the learned latent space.
- 105
- Development of an UAS for post-earthquake disaster surveying and its application in Ms7.0 Lushan Earthquake, Sichuan, China105
Tests showed that the UAS plays an important role in the work of investigating and gathering information about disaster in epicentral areas of the earthquake, such as road detection, secondary disaster investigation, and rapid disaster evaluation.
- Discriminative Spatial Attention for Robust Tracking105
This paper presents an efficient two-stage method that divides the discriminative domain into a local and a semi-local one, and shows that the set of semi- local ARs are identified through an efficient branch-and-bound procedure that guarantees the optimality.
- 104
- 99
- Cooperative Learning of Energy-Based Model and Latent Variable Model via MCMC Teaching94
A cooperative learning algorithm to train both the undirected energy-based model and the directed latent variable model jointly is proposed and it is shown that the cooperativelearning algorithm can learn realistic models of images.
- 93
- Dynamic hand gesture recognition: An exemplar-based approach from motion divergence fields92
A new visual representation for hand motions based on the motion divergence fields, which can be normalized to gray-scale images, and achieves higher recognition accuracy than other state-of-the-art motion and spatio-temporal features on this database.
- Generative PointNet: Deep Energy-Based Learning on Unordered Point Sets for 3D Generation, Reconstruction and Classification91
This model synthesizes point clouds by matching observed examples in terms of statistical properties defined by the energy function, and can learn a short-run MCMC toward the energy-based model as a flow-like generator for point cloud reconstruction and interpolation.
- Generative PointNet: Energy-Based Learning on Unordered Point Sets for 3D Generation, Reconstruction and Classification91
This model synthesizes point clouds by matching observed examples in terms of statistical properties defined by the energy function, and can learn a short-run MCMC toward the energy-based model as a flow-like generator for point cloud reconstruction and interpolation.
- Object Detection with a Unified Label Space from Multiple Datasets88
This work designs a framework which works with such partial annotations, and proposes loss functions that carefully integrate partial but correct annotations with complementary but noisy pseudo labels.
- 88
- 88
- 86
- 85
- 84
- An Adaptive Self-Organizing Color Segmentation Algorithm with Application to Robust Real-time Human Hand Localization83
An adaptive self-organizing color segmentation algorithm and a transductive learning algorithm used to localize human hand in video sequences and color cue and motion cue are integrated in the localization system, in which motion cue is employed to focus the attention of the system.
- Learning Generative ConvNets via Multi-grid Modeling and Sampling82
It is shown that this multi-grid method can learn realistic energy-based generative ConvNet models, and it outperforms the original contrastive divergence (CD) and persistent CD.
- Emergence of analogy from relation learning81
A computational model demonstrating that abstract relations, such as synonymy and antonymy, can be learned efficiently from semantic feature vectors for individual words and can be used to solve simple verbal analogy problems with close to human-level accuracy is developed.
- Monocular Video Foreground/Background Segmentation by Tracking Spatial-Color Gaussian Mixture Models81
- 80
- Modeling Visual Patterns by Intergrating Descriptive and Generative Methods80
This paper demonstrates the integrated framwork by constructing a class of hierarchical models for texton patterns, where the spatial arrangements of the textons in the texton maps are characterized by minimax entropy principle, which leads to embellished versions of Gibbs point rocess.
- Review of Clustering Algorithms79
Clustering can be considered the most important unsupervised learning problem; as with every other problem of this kind, it deals with finding structure in a collection of unlabeled data.
- 77
- Divergence Triangle for Joint Training of Generator Model, Energy-Based Model, and Inferential Model74
The divergence triangle is a compact and symmetric (anti-symmetric) objective function that seamlessly integrates variational learning, adversarial learning, wake-sleep algorithm, and contrastive divergence in a unified probabilistic formulation that makes the processes of sampling, inference, and energy evaluation readily available.
- Growing Interpretable Part Graphs on ConvNets via Multi-Shot Learning74
A learning strategy that embeds object-part concepts into a pre-trained convolutional neural network in an attempt to explore explicit semantics hidden in CNN units and gradually transform the pre- trained CNN into a semantically interpretable graphical model for hierarchical object understanding.
- Interpretable CNNs for Object Classification73
This paper proposes a generic method to learn interpretable convolutional filters in a deep Convolutional neural network for object classification, where each interpretable filter encodes features of a specific object part.
- 73
- 73
- 72
- A sequential quadratic programming algorithm with an additional equality constrained phase72
A sequential quadratic programming (SQP) method is presented that aims to overcome some of the drawbacks of contemporary SQP methods by adding an equality constrained phase that promotes fast convergence and improves performance in the presence of ill conditioning.
- Visual Agents as Fast and Slow Thinkers70
This work introduces FaST, which incorporates the Fast and Slow Thinking mechanism into visual agents, and advocates a flexible system, hierarchical reasoning capabilities, and a transparent decision-making pipeline, all of which contribute to its ability to emulate human-like cognitive processes in visual intelligence.
- 70
- 70
- 70
- Regularized Simultaneous Forward–Backward Greedy Algorithm for Sparse Unmixing of Hyperspectral Data69
- Robust Transfer Learning with Pretrained Language Models through Adapters66
This work inserts small bottleneck layers (i.e., adapter) within each layer of a pretrained model, then fix the pretrained layers and train the adapter layers on the downstream task data, leading to improved stability and adversarial robustness in transfer learning to various downstream tasks.
- 65
- 65
- AdaBoost-based face detection for embedded systems64
This paper designs an embedded face detection system for handheld digital cameras or camera phones that can achieve about 75-80% detection rate for group portraits and proposes a hardware pipeline design for Haar-like feature calculation and a system design exploiting several levels of parallelism.
- Joint Training of Variational Auto-Encoder and Latent Energy-Based Model63
This paper proposes a joint training method to learn both the variational auto-encoder (VAE) and the latent energy-based model (EBM), and shows that the joint training greatly improves the synthesis quality of the VAE.
- 63
- 62
- 59
- Beyond Pass@1: Self-Play with Variational Problem Synthesis Sustains RLVR58
This paper proposes an online Self-play with Variational problem Synthesis (SvS) strategy for RLVR training, which uses the policy's correct solutions to synthesize variational problems while ensuring their reference answers remain identical to the originals.
- Mobile Product Image Search by Automatic Query Object Extraction58
A weighted object mask voting algorithm is proposed based on a spatially-constrained model, which allows robust localization and segmentation of the query object, and achieves significantly better retrieval accuracy than previous methods.
- 58
- Decoding the encoding of functional brain networks: An fMRI classification comparison of non-negative matrix factorization (NMF), independent component analysis (ICA), and sparse coding algorithms57
The success of sparse coding algorithms suggests that algorithms which enforce sparsity, discourage multitasking, and promote local specialization may capture better the underlying source processes than those which allow inexhaustible local processes such as ICA.
- 57
- 57
- Learning Energy-Based Spatial-Temporal Generative ConvNets for Dynamic Patterns56
It is shown that an energy-based spatial-temporal generative ConvNet can be used to model and synthesize dynamic patterns and that the learning algorithm can synthesize realistic dynamic patterns.
- 56
- 54
- From frequent itemsets to semantically meaningful visual patterns54
The novelty of this work lies in the following new contributions: a principled solution to the discovery of meaningful itemsets based on frequent itemset mining; a self-supervised clustering scheme of the high-dimensional visual features by feeding back discovered patterns to tune the similarity measure through metric learning; and a pattern summarization method that deals with the measurement noises brought by the image data.
- 51
- 50
- Learning Multi-layer Latent Variable Model via Variational Optimization of Short Run MCMC for Approximate Inference50
Noise initialized non-persistent short run MCMC, such as finite step Langevin dynamics initialized from the prior distribution of the latent variables, are used as an approximate inference engine and show that the proposed method outperforms variational auto-encoder (VAE) in terms of reconstruction error and synthesis quality.
- Mining Motifs from Human Motion50
This paper proposes an efficient motif discovery method which can handle both spatia l and temporal variations of motion data, and uses locality-sensitive hashing (LSH) to find the approximate matches in a matching trellis.
- 50
- Learning Sparse FRAME Models for Natural Image Patterns49
The experiments show that the sparse FRAME models are capable of representing a wide variety of object patterns in natural images and that the learned models are useful for object classification.
- From information scaling of natural images to regimes of statistical models49
This work shows that the entropy rate of the image data changes over the viewing distance (as well as the camera resolution), and proposes a full-zoom primal sketch model that integrates both sparse coding and Markov random fields.
- 49
- Learning Dynamic Generator Model by Alternating Back-Propagation through Time48
This paper shows that the dynamic generator model for spatial-temporal processes such as dynamic textures and action sequences in video data can be learned by an alternating back-propagation through time algorithm that iteratively samples the noise vectors and updates the parameters in the transition model and the generator model.
- 48
- 46
- 46
- 44
- 42
- 41
- Extraction of an Explanatory Graph to Interpret a CNN40
The explanatory graph transferred features of object parts to the task of part localization, and the method significantly outperformed other approaches.
- 40
- 39
- 37
- 36
- 35
- 34
- 34
- 34
- 33
- Exploiting Vector Fields for Geometric Rectification of Distorted Document Images32
A segment-free method for geometric rectification of a distorted document image captured by a hand-held camera that does not have to extract curved text lines or textual blocks and can be freely applied to document images with extremely complicated page layouts and severe image quality degradation.
- 31
- Mining GPS traces and visual words for event classification31
Experimental results using leave-one-out cross validation have shown the synergy of all of the components in the proposed approach to event classification, which performs multiclass AdaBoost to solve the multiclass problem.
- Digital Twin-based Quality Management Method for the Assembly Process of Aerospace Products with the Grey-Markov Model and Apriori Algorithm30
A DT-based quality management system for the assembly process of aerospace products is developed, which can effectively improve the efficiency of quality management for the Assembly process of Aerospace products and reduce quality abnormalities.
- Interactive visual object search through mutual information maximization30
This work forms the object search as the problem of finding subimages with the largest mutual information toward the query object, and proposes a discriminative matching using both positive and negative queries to obtain the mutual information score.
- 29
- On Learning Non-Convergent Short-Run MCMC Toward Energy-Based Model.29
It is shown that the learned short-run MCMC is capable of generating realistic images, unlike traditional EBM or MCMC, and is also capable of reconstructing observed images and interpolating different images, like generator model or flow model.
- 27
- 27
- 26
- Mining Object Parts from CNNs via Active Question-Answering26
This paper uses active question-answering to semanticize neural patterns in conv-layers of the CNN and mine part concepts to construct an And-Or graph (AOG) to represent a four-layer semantic hierarchy of the part.
- Automatic, fast, online calibration between depth and color cameras26
This work develops a completely automatic, very fast, online algorithm that demonstrates how a consumer-grade depth camera can be calibrated with a color camera with minimal user interaction.
- Exploring Texture Ensembles by Efficient Markov Chain Monte Carlo26
A n efficient M a r k o v chain M o n t e C a r l o a l g o r i t h m is proposed for s a m p l i n g Julesz ensembles, establishing the intrinsic link between the scientific definition of texture o n Z models of texture o n finite lattices.
- 25
- 24
- Energy-Based Continuous Inverse Optimal Control23
This article study the problem of continuous inverse optimal control in the framework of energy-based model (EBM), where the observed expert trajectories are assumed to be random samples from a probability density function defined as the exponential of the negative cost function up to a normalizing constant.
- Conformal Isometry of Lie Group Representation in Recurrent Network of Grid Cells23
This paper focuses on a simple non-linear recurrent model that underlies the continuous attractor neural networks of grid cells and shows that conformal isometry leads to hexagon periodic patterns in the grid cell responses and the model is capable of accurate path integration.
- Inducing wavelets into random fields via generative boosting23
This paper proposes a learning algorithm for the random field models whose energy functions are in the form of linear combinations of rectified filter responses from subsets of wavelets selected from a given over-complete dictionary and demonstrates that the models can be learned in an unsupervised setting.
- Unsupervised learning of compositional sparse code for natural image representation23
Experiments show that the proposed unsupervised learning method is capable of learning meaningful compositional sparse code, and the learned templates are useful for image classification.
- Optimization Strategies for Self-Supervised Learning in the Use of Unlabeled Data22
A novel method is proposed that significantly enhances the performance of algorithms on unlabeled data, achieving improved accuracy and generalization capabilities and is validated across multiple datasets, demonstrating superior performance compared to traditional approaches.
- Recognizing Part Attributes With Insufficient Data22
A novel Concept Sharing Network (CSN) is introduced that has the capability of recognizing the part attribute that has insufficient or zero training data, by learning the part location and appearance pattern respectively from the training data that usually mix them in a single label.
- Hierarchical Dynamic Parsing and Encoding for Action Recognition22
This paper describes a novel hierarchical dynamic encoding method to capture both the locally smooth dynamics and globally drastic dynamic changes and provides a multi-layer joint representation for temporal modeling for action recognition.
- Closed-Loop Adaptation for Robust Tracking22
This work proposes a novel closed-loop model adaptation framework based on the combination of matting and tracking that largely avoids model drift and significantly outperforms other discriminative tracking models as well as video matting approaches.
- SocAoG: Incremental Graph Parsing for Social Relation Inference in Dialogues21
This work model the social network as an And-or Graph, named SocAoG, for the consistency of relations among a group and leveraging attributes as inference cues, and proposes an incremental structure prediction task.
- 21
- 21
- Self-Supervised Learning for Visual Tracking and Recognition of Human Hand21
This paper proposes a new learning paradigm, selfsupervised learning, to investigate the issues of learning bootstrapping and model transduction, and the proposed algorithm, Discriminant-EM (D-EM), is a specific learning technique.
- An empirical study of the effect of background data size on the stability of SHapley Additive exPlanations (SHAP) for deep learning models20
This empirical study on the MIMIC-III dataset shows that the two core explanations - SHAP values and variable rankings fluctuate when using different background datasets acquired from random sampling, indicating that users cannot unquestioningly trust the one-shot interpretation from SHAP.
- Bayesian variable selection for finite mixture model of linear regressions20
A Bayesian variable selection method for fitting the finite mixture model of linear regressions where two sets of latent variables are introduced, indicating which sub-population each observation comes from and whether or not a variable is included in the regression model of a sub- Population.
- 20
- 20
- For Vision-Based Human Computer Interaction20
Since rich visual information provides a strong cue to infer the inner states of an object, vision-based techniques provide promising alternatives to capture human hand motion, and vision systems could be very cost efficient and noninvasive.
- 19
- 19
- 19
- Speeding up spatio-temporal sliding-window search for efficient event detection in crowded videos19
This work proposes a heuristic branch-and-bound solution for event detection in videos by directly finding subvolumes whose scores are higher than the threshold, which dramatically reduces computational cost without significantly degrading the quality of the detection results.
- Statistical Modeling of Texture Sketch19
This work proposes a two-level generative model for textures that consists of a Markov model for the placement of the image bases or the sketch, and the model is characterized by a set of simple geometrical feature statistics.
- Representation Learning: A Statistical Perspective18
Recent advances in learning representations from a statistical perspective are reviewed, including unsupervised learning of vector representations and learning of both vector and matrix representations.
- Learning distance for sequences by learning a ground metric18
The metric is formulated as a parameter of the distance, the objective is that the distance induced by the learned ground metric produces large values for sequences from different classes and small values for those from the same class.
- 18
- A tale of three probabilistic families: Discriminative, descriptive, and generative models18
Three families of probability models, namely, the discriminative models, the descriptive models, and the generative models are reviewed, which take advantage of the high approximation capacities of deep neural networks.
- 18
- Bootstrap Initialization of Nonparametric Texture Models for Tracking18
It is found that (1) independence assumptions in the learning process can be violated to a significant degree, if enough data is taken; (2) there are both domain-independent and domain-specific means to mitigate learning bias; and (3) repeated bootstrapping does not necessarily result in increasingly better models.
- Learning Energy-based Model with Flow-based Backbone by Neural Transport MCMC17
The model has a particularly simple form in the space of the latent variables of the flow-based model, and MCMC sampling of the EBM in the latent space, which is a simple special case of neural transport MCMC, mixes well and traverses modes in the data space.
- 17
- 17
- Self-supervised learning based on discriminative nonlinear features for image classification17
KDEM is extended to a nonlinear kernel algorithm, Kernel DEM (KDEM), and evaluates KDEM extensively on benchmark image databases and synthetic data, showing the effectiveness of the approach.
- Interactive and Incremental Learning via a Mixture of Supervised and Unsupervised Learning Strategies17
A novel algorithm is proposed that can overcome some of existing problems through a mixture of these two paradigms of machine learning, and is tested with a robot language-learning task.
- 17
- Balancing Between Forgetting and Acquisition in Incremental Subpopulation Learning16
A novel two-stage learning scheme to explicitly disentangle the acquisition and forgetting for achieving a better balance between sub-population learning and seen population forgetting is proposed and significantly outperform other methods by a large margin.
- Neural-Symbolic Recursive Machine for Systematic Generalization16
The findings demonstrate that NSR's design, imbued with the inductive biases of equivariance and compositionality, grants it the expressiveness to adeptly handle diverse sequence-to-sequence tasks and achieve unparalleled systematic generalization.
- 16
- 16
- Latent Plan Transformer for Trajectory Abstraction: Planning as Latent Space Inference15
This work introduces the Latent Plan Transformer (LPT), a novel model that leverages a latent variable to connect a Transformer-based trajectory generator and the final return, and validate that latent variable inference can be a strong alternative to step-wise reward prompting.
- 15
- 15
- 15
- 15
- 15
- FedVAE: Trajectory privacy preserving based on Federated Variational AutoEncoder14
A Federated Variational AutoEncoder (FedVAE) approach is proposed, which effectively generates a new trajectory dataset while preserving the confidentiality of private information and retaining the structure of the original features.
- Learning Cycle-Consistent Cooperative Networks via Alternating MCMC Teaching for Unsupervised Cross-Domain Translation14
A generative framework, in which the probability distribution of each domain is represented by a generative cooperative network that consists of an energy-based model and a latent variable model, which is useful for unsupervised image-to-image translation and unpaired image sequence translation.
- Mining Interpretable AOG Representations From Convolutional Networks via Active Question Answering14
A question-answering method that uses active human-computer communications to mine patterns from a pre-trained CNN, in order to incrementally explain more features in conv-layers, and achieves similar or better part-localization performance than fast-RCNN methods.
- 14
- Semiparametric Bayesian analysis of transformation linear mixed models14
A semiparametric Bayesian approach is developed to make Bayesian inference on the novel LMM by using the truncated centered Dirichlet Process prior to approximate the unknown distribution of the random effects and using Bayesian P-splines to approximateThe transformation function.
- 14
- A decentralized probabilistic approach to articulated body tracking14
A novel decentralized probabilistic approach to visual tracking of articulated objects based on a dynamic Markov network, a generative model which characterizes the dynamics, the image observations of each individual limb, as well as the spatial coherence among them is presented.
- Mining repetitive clips through finding continuous paths14
A novel method which translates repetitive clip mining to the continuous path finding problem in a matching trellis, where sequence matching can be accelerated by taking advantage of the temporal redundancies in the videos.
- 14
- Sequential mean field variational analysis of structured deformable shapes14
A novel approach to analyzing and tracking the motion of structured deformable shapes, which consist of multiple correlated deformable subparts, is proposed, which achieves very efficient Bayesian inference of the structured deformation with close-to-linear complexity.
- Learning Joint Latent Space EBM Prior Model for Multi-layer Generator13
This paper proposes an energy-based model (EBM) on the joint latent space over all layers of latent variables with the multi-layer generator as its backbone and develops a joint training scheme via maximum likelihood estimation (MLE), which involves Markov Chain Monte Carlo sampling for both prior and posterior distributions of the latent variables from different layers.
- Discriminative Transformation for Multi-dimensional Temporal Sequences13
A method to transform features in sequences into a low-dimensional subspace such that different sequence classes are holistically separated and provides a new tractable and effective solution with theoretical proofs by constraints unfolding and pruning, convex relaxation, and within-class scatter compression.
- Watermarking Generative Tabular Data12
The proposed watermark can be effectively detected, while faithfully preserving the data fidelity, and also demonstrates appealing robustness against additive noise attack, highlighting its utility in enhancing the security of synthetic and real-world datasets.
- Learning Hierarchical Features with Joint Latent Space Energy-Based Prior12
This paper proposes a joint latent space EBM prior model with multi-layer latent variables for effective hierarchical representation learning and develops a variational joint learning scheme that seamlessly integrates an inference model for efficient inference.
- 12
- 12
- 12
- From an ordinary paper to a wireless and mobile input device12
This paper presents a vision-based gesture interface system, VISUAL PANEL, which employs an arbitrary quadrangle-shaped panel and a tip pointer as an intuitive input device that can accurately and reliably track the panel and the tip pointer.
- VISUAL PANEL: From an ordinary paper to a wireless and mobile input device12
This paper presents a vision-based gesture interface system, VISUAL PANEL, which employs an arbitrary quadrangle-shaped panel and a tip pointer as an intuitive input device that can accurately and reliably track the panel and the tip pointer.
- Learning Neural Representation of Camera Pose with Matrix Representation of Pose Shift via View Synthesis11
This work proposes an approach to learn neural representation of camera poses and 3D scenes, coupled with neural representations of local camera movements, represented as vectors and the local camera movement is represented as a matrix operating on the vector of the camera pose.
- Discussion of "The Art of Data Augmentation" by van Dyk and Meng11
These thoughts (or rather speculations) on the problem of vision from the perspective of missing data modeling and data augmentation are discussed.
- Flexible Visual Recognition by Evidential Modeling of Confusion and Ignorance10
The effectiveness of the methods are demonstrated in quantifying two sources of uncertainties and dealing with flexible recognition through a series of experiments on synthetic data analysis, visual recognition, and open-set detection.
- Classification of Glomerular Pathology Images in Children Using Convolutional Neural Networks with Improved SE-ResNet Module10
Experimental results showed that the classification algorithm proposed in the present study is able to identify glomerular lesions with a higher precision, and distinguish similar glomerular pathologies from each other.
- Diverse and Faithful Knowledge-Grounded Dialogue Generation via Sequential Posterior Inference10
This paper presents an end-to-end learning framework, termed Sequential Posterior Inference (SPI), capable of selecting knowledge and generating dialogues by approximately sampling from the posterior distribution, and demonstrates that SPI outperforms previous strong baselines according to both automatic and human evaluation metrics.
- Variational Bayesian partially linear mean shift models for high‐dimensional Alzheimer's disease neuroimaging data10
A variational Bayesian inference is developed to simultaneously estimate parameters and nonparametric functions and identify outliers in a partially linear mean shift model (PLMSM) to investigate the relationship between MMSE score and high‐dimensional regions of interest in MRI, and detect the outliers.
- 10
- 10
- 10
- 10
- Synthesizing Dynamic Textures and Sounds by Spatial-Temporal Generative ConvNet.10
This paper studies the problem of modeling and synthesizing dynamic textures using a generative version of the convolution neural network (ConvNet or CNN) that consists of multiple layers of spatial-temporal filters to capture the spatial-Temporal patterns in the dynamic textures and shows that such spatial- Temporal generative ConvNet can synthesize realistic dynamic textures.
- 10
- 10
- 10
- 10
- 9
- 9
- No Head Left Behind – Multi-Head Alignment Distillation for Transformers9
A new Attention Map Alignment Distillation (AMAD) method for Transformers with multi-head attention, which works for a Teacher and a Student with different numbers of attention heads, and outperforms conventional distillation methods on VQA-2.0, COCO captioning, and Multi30K translation datasets.
- Molecule Design by Latent Space Energy-Based Modeling and Gradual Distribution Shifting9
A probabilistic generative model to capture the joint distribution of molecules and their properties and a sampling with gradual distribution shifting (SGDS) algorithm, so that after learning the model initially on the training data of existing molecules, the proposed algorithm gradually shifts the model distribution towards the region supported by molecules with desired values of properties.
- 9
- Learning Low-Dimensional Temporal Representations9
This work jointly learns the subspace and the associated alignments by optimizing an objective which favors easily-separable temporal structures, and shows that this objective is connected to the inference of alignments, thus allows an iterative solution.
- 9
- A novel unsupervised approach to discovering regions of interest in traffic images9
The study shows that the distribution of such data can be captured by a simplex in a linear subspace, and each data point can be represented by a linear reconstruction over the set of vertices of the simplex.
- 9
- A null space method for over-complete blind source separation9
A null space representation for the over-complete blind source separation problem is proposed and the problem can be posed in the framework of Bayesian latent variable model, where the mixing matrix and the source sounds can be inferred based on their posterior distributions.
- 9
- 9
- Towards self-exploring discriminating features for visual learning9
A novel method is proposed, the Discriminant-EM (D-EM) algorithm, which attacks these diculties by integrating discriminant analysis with the EM framework in this hybrid formulation, and provides better ability to simplify the probabilistic structures of the visual learning tasks.
- Learning mixtures of Bernoulli templates by two-round EM with performance guarantee8
It is shown that the two-round EM algorithm can learn mixture of Bernoulli templates with near optimal precision with high probability, if the Bernouelli templates are sufficiently different and if the number of features is sufficiently high.
- Linear programming, recurrent associative memories, and feed-forward neural networks8
This paper compares the performance of a recurrent associative memory to that of a feed-forward neural network trained with the same data and finds that the neural network's performance is much less promising than that of the associativeMemory.
- 7
- 7
- SCRIPT: Self-Critic PreTraining of Transformers7
Self-CRItic Pretraining Transformers (SCRIPT) is introduced for representation learning of text and improved sample-efficiency in pretraining and enhanced representations evidenced by improved downstream task performance on GLUE and SQuAD over strong baselines.
- 7
- 7
- 7
- 7
- Context region discovery for automatic motion compensation in fluoroscopy7
An automatic motion compensation scheme to boost image guidence power in transcatheter aortic valve implantation (TAVI) without user interventions is proposed, specially in case of severe occlusion where conventional tracking methods are challenged.
- Single image super-resolution based on space structure learning7
A new SR method is proposed that identifies a space from the low-resolution image space that best preserves the structure of the high-resolution (HR) image space and the inference between the two structure-consistent spaces proves to be accurate and predicts HR image patches with higher quality.
- MOTION SEGMENTATION BASED ON INDEPENDENT SUBSPACE ANALYSIS7
The geometrical essence of the method is regarded as a global constraint added in the segmentation process resulting in a considerable increase in error tolerance, without either prior knowledge of the number of objects or prior assumption about existence of degeneracy.
- From text to insight: A natural language processing-based analysis of burst and research trends in HER2-low breast cancer patients6
Six main research directions are revealed: HER2 detection, omics and clinical biomarkers, basic and translational research, neoadjuvant therapy and prognosis, progress of ADC drugs and clinical trials, and therapeutic efficacy and safety of antibodydrug conjugates (ADCs).
- Monitoring Primitive Interactions During the Training of DNNs6
It is shown that the seemingly complex learning dynamics of a DNN can be faithfully represented as the change of a few primitive interaction patterns encoded by the DNN, and the interaction of principal feature components in intermediate-layer features is defined.
- 6
- 6
- 6
- Emotional Voice Conversion with Semi-Supervised Generative Modeling6
This paper demonstrates that us-ing as little as 1% supervised data is sufficient to achieve EVC, and proposes SGEVC, a novel semi-supervised generative model for emotional voice conversion that achieves state-of-the-art performance and consistently outperforms EVC baseline frameworks.
- On the Complexity of Bayesian Generalization6
Experimental results show that representations with relatively high subjective complexity outperform those with relatively low subjective complexity in the rule- based generalization, while the trend is the opposite in the similarity-based generalization.
- 6
- 6
- Context-aware and attentional visual object tracking6
This thesis presents an in-depth analysis of the chicken-and-egg nature of on-line adaptation of target observation models directly using the previous tracking results, and proposes two novel ideas to combat unpredictable variations: context-aware tracking and attentional tracking.
- Vision and learning for intelligent human-computer interaction6
This dissertation presents three effective techniques for visual motion analysis tasks: non-stationary color model adaptation for efficient localization, multiple visual cues integration for robust tracking, and learning motion models for capturing articulated hand motion.
- 5
- 5
- Unlocking the Potential of Text-to-Image Diffusion with PAC-Bayesian Theory5
This work proposes a Bayesian approach that designs custom priors over attention distributions to enforce desirable properties, including divergence between objects, alignment between modifiers and their corresponding nouns, minimal attention to irrelevant tokens, and regularization for better generalization.
- 5
- 5
- Generative Text Modeling through Short Run Inference5
It is shown that the models trained with short run dynamics more accurately model the data, compared to strong language model and VAE baselines, and exhibit no sign of posterior collapse, and a well-structured latent space of the generative model is exposed.
- 5
- Active target tracking: A simplified view aligning method for binocular camera model5
A novel view aligning method is presented which introduces three reasonable and mild simplifications to transform the complex view alignment problem into a simple arctangent control function and outperforms other state-of-the-art methods on accuracy and effectiveness for active target tracking tasks.
- 5
- 5
- Learning spatio-temporal dependency of local patches for complex motion segmentation5
This work presents a novel method for motion segmentation by learning the motion priors from exemplar motions to guide the segmentation, which is computationally more efficient and able to better handle the inaccurate inference of pairwise relationships.
- 5
- 5
- 4
- 4
- Research on Complaint Operation Management System Based on digital Transformation4
This paper takes the digital transformation of China Telecom customer service as an example, and puts forward the implementation plan of digital complaint operation management, focusing on the construction of digital service capability of complaint acceptance, return visit, archiving, analysis and supervision modules.
- 4
- 4
- A Study of Local Optima for Learning Feature Interactions using Neural Networks4
This paper studies the datastarved regime where a NN is trained on a relatively small amount of training data, and proposes a node pruning and feature selection algorithm that improves the capability of NNs to find better local minima even when there are irrelevant variables.
- 4
- 4
- A Representational Model of Grid Cells Based on Matrix Lie Algebras.4
By reconstructing the kernel functions for place cells, the model learns hexagon grid patterns that characterize the grid cells, and the learned model is capable of near perfect path integral, and it is also capable of error correction.
- 4
- Multimodal Conditional Learning with Fast Thinking Policy-like Model and Slow Thinking Planner-like Model.4
This paper solves the supervised learning of the conditional distribution of a high-dimensional output given an input, where the output and input belong to two different modalities, e.g., the output is an image and the input is a sketch by learning two models that bear similarities to those in reinforcement learning and optimal control.
- Replicating Neuroscience Observations on ML/MF and AM Face Patches by Deep Generative Model4
The generator network is learned from the face images generated by a pretrained AAM model using a variational autoencoder, and it is shown that the inferred latent variables of the learned generator network have a strong linear relationship with the shape and appearance variables of this model.
- Learning Grid-like Units with Vector Representation of Self-Position and Matrix Representation of Self-Motion.4
A model for learning grid-like units for spatial awareness and navigation by minimizing a loss function that combines the loss functions of the two sub-models, which can be used for path integral and path planning.
- 4
- 4
- Learning Generative ConvNet with Continuous Latent Factors by Alternating Back-Propagation.4
This paper shows that the unsupervised learning of a popular top-down generative ConvNet model with latent continuous factors can be accomplished by a learning algorithm that consists of alternatively performing back-propagation on both the latent factors and the parameters.
- Learning FRAME Models Using CNN Filters for Knowledge Visualization4
This paper proposes to learn the generative FRAME (Filters, Random field, And Maximum Entropy) model using the highly expressive filters pre-learned by the CNN at the convolutional layers, and explains how this model corresponds to a CNN unit at a layer above the layer of filters employed by the model.
- Special Issue on Video Analysis on Resource-Limited Systems4
This special issue seeks to present and highlight the recent developments in the area of video analysis on resource-limited systems.
- Perceptual Scaling4
Vision can be posed as a statistical learning and inference problem and a number of statistical theories proposed for vision, including pattern theory as a paradigm for vision are proposed.
- On the Analysis and Distillation of Emergent Outlier Properties in Pre-trained Language Models3
The Emergent Outlier Focused Distillation (EOFD) method, which prioritizes critical outlier dimensions in distillation using a weighted MSE loss, is proposed and empirically demonstrated that EOFD outperforms state-of-the-art distillation methods and generalizes well across Encoder-only BERT, Decoder-only GPT-2, and Encoder-Decoder T5 architectures.
- DODT: Enhanced Online Decision Transformer Learning through Dreamer's Actor-Critic Trajectory Forecasting3
A novel approach that combines the Dreamer algorithm's ability to generate anticipatory trajectories with the adaptive learning strengths of the Online Decision Transformer to create a bidirectional enhancement loop that accelerates learning and showcases robustness in diverse and dynamic scenarios.
- 3
- 3
- 3
- Inbound Tourists’ Perception of Tourist Destination Image Classified by UGC Picture Computer Program3
The research results show that, in the analysis of specific inbound tourists’ perception of tourist destination image, Shanghai is a very modern cultural city with very distinctive architectural features; inbound visitors from East Asia are more and more interested in the cultural landscape and natural scenery of Shanghai.
- SAS: Self-Augmentation Strategy for Language Model Pre-training3
A self-augmentation strategy (SAS) where a single network is utilized for both regular pre-training and contextualized data augmentation for the training in later epochs, which is able to outperform ELECTRA and other state-of-the-art models in the GLUE tasks with similar or less computation cost.
- A Novel Phoneme-based Modeling for Text-independent Speaker Identification3
Experimental results show that the proposed phoneme-based approach is effective and robust, which out-performs the state-of-the-art methods in both EER and accuracy, especially with a larger phonemic mismatch between the enrollment and test utterances.
- Research on Intelligent Customer Service Application Architecture Based on Digital Transformation3
The typical application of China Telecom intelligent customer service system in digital transformation in detail is introduced and the functional architecture and implementation planning scheme of intelligent application is put forward by optimizing the key modules of intelligent voice navigation.
- 3
- 3
- 3
- 3
- 3
- Learning vector representation of local content and matrix representation of local motion, with implications for V13
A representational model for image pair such as consecutive video frames that are related by local pixel displacements that can learn to infer local motions and can learn Gabor-like filter pairs of quadrature phases is proposed.
- Mining deep And-Or object structures via cost-sensitive question-answer-based active annotations3
A cost-sensitive active Question-Answering (QA) framework for learning a nine-layer And-Or graph (AOG) from web images that iteratively selects different storylines of questions to update different nodes in the AOG.
- Guest Editors’ Introduction to the Special Section on Learning with Shared Information for Computer Vision and Multimedia Analysis3
The twelve papers in this special section focus on learning systems with shared information for computer vision and multimedia communication analysis, which are very effective in solving real-world large-scale problems.
- Learning Multi-grid Generative ConvNets by Minimal Contrastive Divergence.3
It is shown that this method can learn realistic energy-based generative ConvNet models, and it outperforms the original contrastive divergence (CD) and persistent CD.
- 3
- 3
- Random Occupancy Patterns3
This work develops a semi-local feature called random occupancy pattern (ROP), which employs a novel progressive rejection sampling scheme to effectively explore an extremely large sampling space and utilizes a sparse coding approach to robustly encode these features.
- 3
- Wavelet, active basis, and shape script: a tour in the sparse land3
It is explained that the effort of seeking a common wavelet sparse coding of images from the same object category leads to an active basis model, where the images share the same set of selected wavelet elements, which form a linear basis for representing the images.
- Statistical Principles in Image Modeling3
This review article describes three statistical principles for modeling image patterns: the sparse coding principle, the minimax entropy principle, and the meaningful alignment principle, which correspond to three regimes of composition patterns of Gabor wavelets.
- Collaborative visual tracking of multiple identical targets3
A collaborative inference mechanism where the estimate of each individual target state is not only determined by its own observation and dynamics, but also through the interaction and collaboration with the state estimates of other targets, which finally leads to a competition mechanism that enables different but spatial adjacent targets to compete for the common image observations.
- The motion planning neural circuit in goal-directed navigation as Lie group operator search2
This work links the abstract operator search with biological neural circuits for the first time and finds the abstract group operator search can be implemented by a two-layer feedforward circuit utilizing circuit motifs of connection phase shift, nonlinear activation function, and pooling, similar to Drosophila's goal-directed navigation neural circuits.
- 2
- An Important Pick-and-Pass Gated Refinement Network for Salient Object Detection in Optical Remote Sensing Images2
A novel important pick-and-pass gated refinement network model for SOD in ORSIs, named IP2GRNet, including a detail refinement stage and a feature pick- and-pass refinement stage, which outperforms the state-of-the-art saliency detectors.
- 2
- 2
- 2
- 2
- 2
- Interpretable CNNs2
A generic method to learn interpretable convolutional convolutional filters in a deep convolutional neural network, where each interpretable convolutional encodes features of a specific object part.
- Learning Deep Generative Models with Short Run Inference Dynamics2
This paper proposes to use short run inference dynamics guided by the log-posterior, such as finite-step gradient descent algorithm initialized from the prior distribution of the latent variables, as an approximate sampler of the posterior distribution, where the step size of the gradient descent dynamics is optimized by minimizing the Kullback-Leibler divergence.
- Replicating Active Appearance Model by Generator Network2
The generator network is learned from the face images generated by a pre-trained AAM model using variational auto-encoder, and it is shown that the inferred latent variables of the learned generator network have strong linear relationship with the shape and appearance variable of the AAM models that generates the face stimuli.
- 2
- 2
- 2
- 2
- Non-standard Clustering Criteria2
The K-means algorithm is one of the most frequently used investigatory algorithms in data analysis and attempts to locate K prototypes or means throughout a data set in such a way that the K prototypes best represent the data.
- 2
- 2
- 2
- 2
- Capturing Human Body Motion from Video for Perceptual Interfaces by Sequential Variational MAP2
It is shown, with theoretic guarantee, that the MAP estimate could be asymptotically achieved from a probabilistic variational approach and achieves linear complexity w.r.t. the number of body parts, which greatly relieves the curse-of-dimensionality.
- Side Statistics and Maximum Discriminant Analysis for Real-Time Tracking2
A new technique for tracking based on side statistics and maximum discriminant analysis that can deal with complex environment, changing background, and partial occlusion, and it is real-time and accurate is proposed.
- Towards Self-Exploring Discriminating Features2
Experimental results show that D-EM outperforms a variety of supervised and semi-supervised learning algorithms for many visual learning tasks, such as content-based image retrieval and invariant object recognition.
- Auto-Encoding Variational Bayesian Inference in High-Dimensional Skew-Normal Linear Mixed Models1
The authors propose a high-dimensional skew-normal linear mixed model and develop a novel variational Baysian method that integrates spike-and-slab Lasso priors for simultaneous parameter estimation and variable selection, and propose a variational auto-encoders to extract latent features.
- 1
- Long-range gene expression prediction with token alignment of large language model1
GTA is introduced, which aligns genetic sequence features with natural language tokens, allowing for symbolic reasoning of genomic sequence features via the frozen language model, and learns the regulatory grammar and allows for in-context learning that is not possible with existing models.
- 1
- 1
- 1
- Emergence of Grid-like Representations by Training Recurrent Networks with Conformal Normalization1
This paper proposes a simple and general conformal normalization of the input velocity for the recurrent neural network, so that the local displacement of the position vector in the high-dimensional neural space is proportional to the local displacement of the agent in the 2D physical space, regardless of the direction of the input velocity.
- 1
- ALUM: Adversarial Data Uncertainty Modeling from Latent Model Uncertainty Compensation1
This work proposes a novel method called ALUM to simultaneously handle the model uncertainty and data uncertainty in a unified scheme to facilitate data uncertainty modeling and non-parametric uncertainty estimations to compensate for the insufficiently trained latent model layers.
- 1
- Beyond Empirical Risk Minimization: Local Structure Preserving Regularization for Improving Adversarial Robustness1
This work proposes a novel Local Structure Preserving (LSP) regularization, which aims to preserve the local structure of the input space in the learned embedding space, and shows strong empirical evidence that with or without adversarial training, this method consistently improves the performance of adversarial robustness.
- 1
- 1
- 1
- SpectraNet: Multivariate Forecasting and Imputation under Distribution Shifts and Missing Data1
This work proposes SpectraNet, a novel multivariate time-series forecasting model that dynamically infers a latent space spectral decomposition to capture current temporal dynamics and correlations on the recent observed history.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- Multi-Shot Mining Semantic Part Concepts in CNNs.1
A new learning strategy that incrementally embeds new object-part concepts into a pre-trained convolutional neural network in order to explore explicit semantics for the CNN units and gradually transfer the pre- trained CNN into a “white-box” model for hierarchical object understanding.
- 1
- 1
- 1
- Multimedia Data Indexing1
In this chapter,.the.authors.discuss.the challenges.of.the.users.to.make.them.accessible.in.digital.data.management.and.capabilities.
- 1
- Conquering image imperfectness by priors with applications to image recovery1
This dissertation proposes the edge smoothness prior to remove the jaggy edge artifacts with the help of the SoftCuts metric, and proposes a novel approach for super resolving one single input image.
- Fusion of Multimodal Partial Estimates1
In this new approach, fusion is related to a computational geometry problem of finding the minimum-volume orthotope, and an effective and scalable branch and bound search algorithm is designed to obtain the global optimal solution.
- Image and video data mining1
This dissertation presents a systematic study on image andVideo data mining and concentrates on two fundamental problems: mining interesting patterns from image and video data and discovering discriminative features for visual pattern classification.
- 1
- 1
- –
- Digital analysis of cultural and creative design elements in the greater bay area based on optical imaging and computer vision–
The model is integrated into a prototype system, and user testing is used to evaluate its usability in pattern retrieval, style recommendation, and color matching assistance, providing an engineering technical path for the structured management and design reuse of cultural and creative resources in the Greater Bay Area.
- –
- –
- –
- –
- Analysis of Error Sources in LLM-based Hypothesis Search for Few-Shot Rule Induction–
This work compares an LLM-based hypothesis search framework with direct program generation approaches on few-shot rule induction tasks and shows that hypothesis search achieves performance comparable to humans, while direct program generation falls notably behind.
- –
- –
- –
- –
- Value-Spectrum: Quantifying Preferences of Vision-Language Models via Value Decomposition in Social Media Contexts–
Value-Spectrum is introduced, a novel Visual Question Answering benchmark aimed at assessing VLMs based on Schwartz's value dimensions that capture core human values guiding people's preferences and actions that highlight the potential of Value-Spectrum as a comprehensive evaluation set for tracking VLM preferences in value-based tasks and abilities to simulate diverse personas.
- –
- –
- –
- –
- –
- Inverse halftoning based on sparse representation with boosted dictionary–
An inverse halftone algorithm based on sparse representation with the dictionary learned by two steps: deconvolution and sparse optimization in the transform domain to remove the noise has better PSNR and visual performance than state-of-the-art methods.
- –
- –
- –
- Triple Regression for Camera Agnostic Sim2Real Robot Grasping and Manipulation Tasks–
The Triple Regression Sim2Real framework is introduced, which constructs a real-time digital twin that serves as a replica of reality to simulate and evaluate multiple plans before their execution in real-world scenarios.
- –
- –
- –
- Application of AI technology based on customer service digitalization–
The intelligent transformation of the back-end service support, with system automation as the core, changes the way of relying on manual maintenance and artificial intelligence technology is used to solve the problem tracing and early warning of artificial customer service in the service process.
- –
- –
- –
- –
- –
- –
- –
- –
- Recognition of Benign and Malignant Breast Ultrasound Images Based on Deep Transfer Learning–
The experimental results had shown that the pre-trained ResNet18 model has the best classification performance on breast ultrasound images, which had proved the significant advantages of deep transfer learning in the classification of small samples of medical images.
- –
- –
- –
- –
- –
- –
- –
- Applied Status and Quality of Graphs in Five Academic Medical Journals in Shanghai (Preprint)–
The quality of papers using graphs was not properly controlled, andGraphs are less commonly applied in the five university journals, and Editors and journal quality management should strengthen the quality control of charts in papers.
- Mining a Deep And-OR Object Semantics from Web Images via Cost-Sensitive Question-Answer-Based Active Annotations–
A cost-sensitive active Question-Answering (QA) framework for learning a nine-layer And-Or graph (AOG) from web images that iteratively selects different storylines of questions to update different nodes in the AOG.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Review of Linear Projection Methods–
This section reviews several projection methods in detail for exploratory data analysis that change the basis of the considered data space by projecting the data to a lower dimensional space.
- –
- –
- Probabilistic variational methods for vision based complex motion analysis–
This dissertation proposes a distributed and collaborative probabilistic reasoning framework for complex motion analysis that handles measurement uncertainty, measurement multi-modality, and measurement inconsistency in a principled way, and is also a distributed visual measurement integration framework.
- Multiple motion analysis for intelligent video surveillance–
This dissertation concentrates on the developments of efficient and effective multiple motion analysis techniques that allow automated tracking of multiple targets, which is arguably the most challenging problem and essential component of any intelligent video surveillance systems.
- –
- On the Parallel/Distributed Network Computing Environment for Wu's Method–
In order to get better performance for the implementation of Wu's method, A network environment which could provide parallel/distributed processing is discussed.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-14. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.