Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks253 from public data
- MLESAC: A New Robust Estimator with Application to Estimating Image Geometry2,696
A new robust estimator MLESAC is presented which is a generalization of the RANSAC estimator which adopts the same sampling strategy as RANSac to generate putative solutions, but chooses the solution that maximizes the likelihood rather than just the number of inliers.
- An Embarrassingly Simple Approach to Zero-Shot Learning1,359
This paper describes a zero-shot learning approach that can be implemented in just one line of code, yet it is able to outperform state of the art approaches on standard datasets.
- The StarCraft Multi-Agent Challenge1,324
The StarCraft Multi-Agent Challenge (SMAC), based on the popular real-time strategy game StarCraft II, is proposed as a benchmark problem and an open-source deep multi-agent RL learning framework including state-of-the-art algorithms is opened.
- 1,042
- 513
- 458
- 412
- 394
- 375
- 372
- What, Where and How Many? Combining Object Detectors and CRFs370
A probabilistic framework for reasoning about regions, objects, and their attributes such as object class, location, and spatial extent is presented, which combines results from sliding window detectors, and low-level pixel-based unary and pairwise relations.
- Graph Cut Based Inference with Co-occurrence Statistics366
This paper considers a class of global potentials defined over all variables in the CRF to show how they can be readily optimised using standard graph cut algorithms at little extra expense compared to a standard pairwise field.
- Geometric motion segmentation and model selection350
This paper places the problems of model selection, robust estimation and clustering into a common statistical framework; investigating the use of information criteria and robust mixture models as a principled way for motion segmentation of images.
- 329
- Robust parameterization and computation of the trifocal tensor325
This paper presents all algorithm for computing a maximum likelihood estimate (MLE) of the trifocal tensor and the proposed parameterization is compared to other existing methods.
- Bayesian Model Estimation and Selection for Epipolar Geometry and Generic Manifold Fitting299
This paper explores ways of automating the model selection process with specific emphasis on the least squares problem of fitting manifolds to data points, illustrated with respect to epipolar geometry.
- 278
- 278
- Combining Appearance and Structure from Motion Features for Road Scene Understanding265
A framework for pixel-wise object segmentation of road scenes that combines motion and appearance features that is designed to handle street-level imagery such as that on Google Street View and Microsoft Bing Maps is presented.
- 260
- 232
- 227
- The Landscape of Agentic Reinforcement Learning for LLMs: A Survey221
A comprehensive twofold taxonomy is proposed: one organized around core agentic capabilities, including planning, tool use, memory, reasoning, self-improvement, and perception, and the other around their applications across diverse task domains.
- 220
- 219
- Modelling and Interpretation of Architecture from Several Images215
This paper describes the automatic acquisition of three dimensional architectural models from short image sequences using Bayesian and model based methods and proves the validity of the prior by verifying that plausible buildings are generated under varying conditions.
- 214
- VideoTrace209
The combination of automated and manual reconstruction allows VideoTrace to model parts of the scene not visible, and to succeed in cases where purely automated approaches would fail.
- Posecut: Simultaneous segmentation and 3D pose estimation of humans using dynamic graph-cuts201
The method is completely generic and can be used to segment and infer the pose of any specified rigid, deformable or articulated object.
- 189
- 183
- The Problem of Degeneracy in Structure and Motion Recovery from Uncalibrated Image Sequences183
The aim of this work is the recovery of 3D structure and camera projection matrices for each frame of an uncalibrated image sequence, and investigates two strategies for tackling degeneracies, including a statistical model selection test to identify when degeneracies occur.
- Gaussian Process Latent Variable Models for Human Pose Estimation175
A dynamical model over the latent space is learned which allows us to disambiguate between ambiguous silhouettes by temporal consistency and is easily extended to multiple observation spaces without constraints on type.
- OASIS: Open Agent Social Interaction Simulations with One Million Agents161
OASIS is a generalizable and scalable social media simulator that replicates various social phenomena, including information spreading, group polarization, and herd effects across X and Reddit platforms, and demonstrates OASIS's potential as a powerful tool for studying complex systems in digital environments.
- 160
- Geometrically Constrained Structure from Motion: Points on Planes160
The experimental results indicate that the quality of the reconstruction can be significantly improved by the judicious use of geometric constraints, and the approaches studied include hallucinating extra point matches in planar regions, computing fundamental matrices directly from homographies, and applying coplanarity and other geometric constraints as part of the final bundle adjustment stage.
- GaussCtrl: Multi-View Consistent Text-Driven 3D Gaussian Splatting Editing149
GaussCtrl, a text-driven method to edit a 3D scene reconstructed by the 3D Gaussian Splatting (3DGS), achieves faster editing and better visual results than previous state-of-the-art methods.
- IMPSAC: Synthesis of importance sampling and random sample consensus147
The method is shown that the method is superior to previous single resolution RANSAC-style feature matchers, and encodes the uncertainty by providing an arbitrarily close approximation to the posterior distribution of the two view relation.
- 145
- Locally Linear Support Vector Machines141
This paper proposes a novel locally linear svm classifier with smooth decision boundary and bounded curvature and shows how this model can be optimized in an online fashion by performing stochastic gradient descent with the same convergence guarantees as standard gradient descent method for linear sVM.
- SiamMask: A Framework for Fast Online Object Tracking and Segmentation.140
This article improves the offline training procedure of popular fully-convolutional Siamese approaches by augmenting their losses with a binary segmentation task, and introduces SiamMask, a framework to perform both visual object tracking and video object segmentation, in real-time, with the same simple method.
- Multivariate Relevance Vector Machines for Tracking139
A one-to-many mapping from image features to state space is learned using a set of relevance vector machines, extended to handle multivariate outputs, which reduces the estimation error in clutter compared to shape-context histograms.
- 135
- 133
- Measuring Uncertainty in Graph Cut Solutions – Efficiently Computing Min-marginal Energies Using Dynamic Graph Cuts128
This paper shows how the min-marginals associated with the label assignments in a MRF can be efficiently computed using a new algorithm based on dynamic graph cuts using a fast and novel algorithm to efficiently minimize these energy projections.
- Filter-Based Mean-Field Inference for Random Fields with Higher-Order Terms and Product Label-Spaces127
The techniques are demonstrated on joint stereo and object labelling problems, as well as object class segmentation, showing in addition for joint object-stereo labelling how the method provides an efficient approach to inference in product label-spaces.
- 125
- Box2Seg: Attention Weighted Loss and Discriminative Feature Learning for Weakly Supervised Segmentation120
This work proposes a weakly supervised approach to semantic segmentation using bounding box annotations that achieves state-of-the-art segmentation accuracy on PASCAL VOC 2012 by improving the mIOU metric by 2 .
- 118
- VMem: Consistent Interactive Video Scene Generation with Surfel-Indexed View Memory117
Surfel-Indexed View Memory is introduced, a memory module that remembers past views by indexing them geometrically based on the 3D surface elements (surfels) they have observed, which enables efficient retrieval of the most relevant past views when generating new ones.
- 114
- 113
- 111
- 106
- 104
- 104
- 104
- Open World Entity Segmentation103
This work proposes a CondInst-like fully-convolutional architecture with two novel modules specifically designed to exploit the class-agnostic and non-overlapping requirements of ES, and investigates the feasibility of convolutional center-based representation to segment things and stuffs in a unified manner.
- Deep Multi-Agent Reinforcement Learning for Decentralized Continuous Cooperative Control102
This work introduces Multi-Agent Mujoco, a novel benchmark suite that, unlike StarCraft II, the predominant benchmark environment, applies to continuous robotic control tasks and presents a range of benchmark results on this new suite, including comparing the state-of-the-art actor-critic method MADDPG against two novel variants of existing methods.
- Hand Pose Estimation Using Hierarchical Detection100
This paper presents an analysis of the design of classifiers for use in a hierarchical object recognition approach, where a cascade of classifier is arranged in a tree in order to recognize multiple object classes.
- CLIP as RNN: Segment Countless Visual Concepts without Training Endeavor96
A novel recurrent framework that progressively filters out irrelevant texts and enhances mask quality without training efforts is introduced, which outperforms not only the training-free counter-parts, but also those fine-tuned with millions of data sam-ples, and sets the new state-of-the-art records for both zero-shot semantic and referring segmentation.
- 92
- Extending Pictorial Structures for Object Recognition90
The goal of this paper is to recognize various deformable objects from images by extending the class of generative probabilistic models known as pictorial structures, particularly suited to represent articulated structures.
- Bilateral Functions for Global Motion Modeling87
This paper uses the bilateral domain to reformulate a piecewise smooth constraint as continuous global modeling constraint and demonstrates how the model can reliably obtain large numbers of good quality correspondences over wide baselines, while keeping outliers to a minimum.
- Solving Markov Random Fields using Semi Definite Programming83
In a roof structure a flexible roofing membrane is supported and secured between at least two layers of tensioned flexible elements such as wires or cords.
- Random forests versus Neural Networks — What's best for camera localization?79
The experimental findings show that for scene coordinate regression, traditional NN architectures are superior to test-time efficient RFs and ForestNets, however, this does not translate to final 6D camera pose accuracy where RFsAndForestNets perform slightly better.
- Sample-dependent Adaptive Temperature Scaling for Improved Calibration76
For each input, this work proposes to predict a different temperature value, allowing us to adjust the mismatch between confidence and accuracy at a finer granularity, and test the method on the ResNet50 and WideResNet28-10 architectures, showing that producing per-data-point temperatures improves the expected calibration error across the whole test set.
- An Analysis of Convex Relaxations for MAP Estimation76
An analysis of three approximate algorithms based on convex relaxations and it is proved that despite the flexibility in the form of the constraints/objective function offered by QP and SOCP, the LP-S relaxation strictly dominates and provides a better approximation than the QP-RL and SOCp-MS.
- Pose estimation and tracking using multivariate regression72
An extension of the relevance vector machine (RVM) algorithm to multivariate regression allows the application to the task of estimating the pose of an articulated object from a single camera.
- On partial optimality in multi-label MRFs72
Theoretical properties of the new relaxation of the multi-label MRF are studied and a subclass of problems where the two relaxations coincide are characterized, and several combined algorithms based on the technique are proposed and demonstrate their performance on challenging computer vision problems.
- Hierarchical Part-Based Human Body Pose Estimation71
A new hierarchical part-based pose estimation method for the upper-body that efficiently searches the high dimensional articulation space using a modified Viterbi algorithm for smooth trajectory of articulation between frames.
- Stochastic motion clustering71
This paper presents a new method for motion segmentation that exploits the fact that two views of a rigidly connected 3D point set are linked by the 3×3 Fundamental Matrix which contains all the information on the motion of a given set of point correspondences.
- Solving Inefficiency of Self-supervised Representation Learning68
This paper proposes a novel self-supervised learning framework using a truncated triplet loss that demonstrates the model’s superiority over the latest state-of-the-art methods by a clear margin.
- Holistic, Instance-level Human Parsing67
This work segments the parts of objects at an instance-level, such that each pixel in the image is assigned a part label, as well as the identity of the object it belongs to, and shows how this approach benefits us in obtaining segmentations at coarser granularities as well.
- An Impartial Take to the CNN vs Transformer Robustness Contest66
In this paper, extensive empirical analyses are performed showing that recent state-of-the-art CNNs (particularly, ConvNeXt) can be as robust and reliable or even sometimes more than the current state of theart Transformers.
- 64
- Measuring uncertainty in graph cut solutions63
This paper shows how the min-marginals associated with the label assignments of a random field can be efficiently computed using a new algorithm based on dynamic graph cuts to compute a measure of uncertainty associated with the graph cut solutions.
- Improved Moves for Truncated Convex Models61
This work considers the problem of obtaining an approximate maximum a posteriori estimate of a discrete random field characterized by pairwise potentials that form a truncated convex model and proposes two st-MINCUT based move making algorithms that are faster and faster than previous approaches based on the LP relaxation.
- Improved Moves for Truncated Convex Models61
This work considers the problem of obtaining an approximate maximum a posteriori estimate of a discrete random field characterized by pairwise potentials that form a truncated convex model and proposes two st-MINCUT based move making algorithms that are faster and faster than previous approaches based on the LP relaxation.
- 54
- AN ANALYSIS OF CONVEX RELAXATIONS FOR MAP ESTIMATION OF DISCRETE MRFs53
An analysis of three algorithms based on convex relaxations and it is proved that despite the flexibility in the form of the constraints/objective function offered by QP and SOCP, the LP-S relaxation strictly dominates and provides a better approximation than QP-RL and SOCP-MS.
- Learning Class-Specific Edges for Object Detection and Segmentation52
It is shown how, for certain object classes, small regions around edges can be used to classify the edge into object or non-object, and performance of both algorithms (matching and segmentation) is considerably improved by the class-specific edge labelling.
- Language Matters: A Weakly Supervised Vision-Language Pre-training Approach for Scene Text Detection and Spotting51
A weakly supervised pre-training method, oCLIP, which can acquire effective scene text representations by jointly learning and aligning visual and textual information and outperforms existing pre- training techniques consistently across multiple public datasets.
- 51
- Dense stereo using pivoted dynamic programming51
This paper presents a unified dynamic programming/statistical framework that allows the incorporation of any partial knowledge about disparities, such as matched features and known surfaces within the scene, with a faster run time and greater accuracy than the standard dynamic programming method.
- When LLMs step into the 3D World: A Survey and Meta-Analysis of 3D Tasks via Multi-modal Large Language Models49
This survey provides a comprehensive overview of the methodologies enabling LLMs to process, understand, and generate 3D data, and examines their integration with LLMs for tasks such as 3D scene understanding, captioning, question-answering, and dialogue.
- 49
- 48
- Learning a Kinematic Prior for Tree-Based Filtering48
The aim in this paper is to track articulated hand motion from monocular video by using a tree-based representation of the posterior distribution using a hierarchical clustering algorithm and two techniques for constructing the tree are described.
- 47
- PoisonBench: Assessing Large Language Model Vulnerability to Data Poisoning44
PoisonBench is introduced, a benchmark for evaluating large language models' susceptibility to data poisoning during preference learning, exposing weaknesses in current preference learning techniques, highlighting the urgent need for more robust defenses against malicious models and data manipulation.
- Learning Layered Pictorial Structures from Video.41
A new unsupervised learning method to obtain a layered pictorial structure (LPS) representation of an articulated object from video sequences is proposed, which has the advantage of achieving a strong local minimum over labels.
- Semantics-Aware Dynamic Localization and Refinement for Referring Image Segmentation39
This work introduces a simple yet effective alternative for progressively learning discriminative multi-modal features by leveraging a continuously updated query as the representation of the target object and at each iteration, strengthen multi- modal features strongly correlated to the query while weakening less related ones.
- Learning Regional Attraction for Line Segment Detection39
The problem of line segment detection (LSD) as a problem of region coloring is posed, and an end-to-end framework to learn attraction field maps for raw input images, followed by a squeeze module to detect line segments is developed.
- Hadamard Product in Deep Learning: Introduction, Advances and Challenges38
This analysis reveals the Hadamard product as a versatile primitive that offers compelling trade-offs between computational efficiency and representational power, positioning it as a crucial component in the deep learning toolkit.
- Reconstructing relief surfaces37
This paper generalizes Markov Random Field stereo methods to the generation of surface relief (height) fields rather than disparity or depth maps, which enables the reconstruction of complete object models using the same algorithms that have been previously used to compute depth maps in binocular stereo.
- Image Processing Issues in Digital Strain Mapping35
It is shown how a probability-based motion detection algorithm can be used as a high confidence estimator of the strain tensor characterising the deformation of the material.
- Bi-Factorial Preference Optimization: Balancing Safety-Helpfulness in Language Models34
A supervised learning framework called Bi-Factorial Preference Optimization is proposed, which re-parameterizes a joint RLHF objective of both safety and helpfulness into a single supervised learning objective and significantly outperforms existing approaches in both safety and helpfulness.
- Tem-adapter: Adapting Image-Text Pretraining for Video Question Answer34
The proposed Tem-adapter enables the learning of temporal dynamics and complex semantics by a visual Temporal Aligner and a textual Semantic Aligner that first designs a template to fuse question and answer pairs as event descriptions and then learns a Transformer decoder with the whole video sequence as guidance for refinement.
- Revisiting Evaluation Metrics for Semantic Segmentation: Optimization and Evaluation of Fine-grained Intersection over Union34
This work proposes the use of fine-grained mIoUs along with corresponding worst-case metrics, thereby offering a more holistic evaluation of segmentation techniques and identifies the crucial role played by architecture designs and loss functions, which lead to best practices in optimizing fine- grained metrics.
- Regression-Based Human Motion Capture From Voxel Data34
A regression based method is proposed to recover human body pose from 3D voxel data using a Bayesian approach based on Mixture of Probabilistic PCA that transforms a collection of 3D shape context descriptors, extracted from the voxels, to a compact feature vector.
- Fields of Experts for Image-based Rendering34
A new optimization algorithm from statistical physics is introduced to computer vision for the first time which reaches better minima than the ICM and simulated annealing algorithms to which such large-clique problems have previously been restricted.
- Rethinking Data Protection in the (Generative) Artificial Intelligence Era32
This work proposes a four-level taxonomy, including non-usability, privacy preservation, traceability, and deletability, that captures the diverse protection needs arising in modern (generative) AI models and systems, and offers a structured understanding of the trade-offs between data utility and control.
- 32
- 31
- Corrective Machine Unlearning30
It is demonstrated that the problem of corrective unlearning has significantly different requirements from traditional privacy-oriented unlearning, and most existing unlearning methods, including retraining-from-scratch without the deletion set, require most of the manipulated data to be identified for effective corrective unlearning.
- Efficiently solving convex relaxations for MAP estimation29
This work builds on the tree reweighted message passing (TRW) framework and shows how the dual formulation of TRW can be extended to include cycle inequalities and SOC constraints, and proposes efficient iterative algorithms for solving the resulting duals.
- Estimating 3D hand pose using hierarchical multi-label classification29
Given a parametric 3D model, generating training data in the form of example images is cheap, and it is demonstrated that it can be used to design classifiers almost as good as those trained using non-synthetic data.
- Robust Vision29
A system—consisting of feature detector, feature matcher, estimation of the Fundamental Matrix, and estimation of structure—that emphasizes robustness to outliers at each stage is presented, and results are shown for the computation of structure from motion in a cluttered and unknown environment.
- WildVidFit: Video Virtual Try-On in the Wild via Image-Based Controlled Diffusion Models28
The solution, WildVidFit, employs image-based controlled diffusion models for a streamlined, one-stage approach that leverages diffusion guidance from pre-trained models including a video masked autoencoder for segment smoothness improvement and a self-supervised model for feature alignment of adjacent frame in the latent space.
- 28
- Dynamic Graph Cuts and Their Applications in Computer Vision28
This chapter addresses the problem of efficient and exact minimization of groups of similar functions which are known to be solvable in polynomial time and presents a novel dynamic algorithm for minimizing such functions.
- Secret Collusion among AI Agents: Multi-Agent Deception via Steganography27
This paper comprehensively formalises the problem of secret collusion in systems of generative AI agents by drawing on relevant concepts from both the AI and security literature, and provides extensive empirical results across a range of contemporary LLMs.
- Local and Global GANs with Semantic-Aware Upsampling for Image Generation27
A local class-specific generative network using semantic maps as guidance, which separately constructs and learns subgenerators for different classes, enabling it to capture finer details of small objects and detailed local textures.
- Three Birds One Stone: A Unified Framework for Salient Object Segmentation, Edge Detection and Skeleton Extraction.27
This paper introduces a horizontal cascade, each component of which is densely connected to the outputs of previous component, which allows for effectively exploiting features across different levels hierarchically to effectively address the multiple pixel-wise binary regression tasks.
- Graph Cuts for Minimizing Robust Higher Order Potentials27
This paper proves that the optimal swap and expansion moves for energy functions composed of these potentials can be computed by solving a st-mincut problem and is able to produce the globally optimal solution in polynomial time.
- Jailbreak-AudioBench: In-Depth Evaluation and Analysis of Jailbreak Threats for Large Audio Language Models26
Jailbreak-AudioBench establishes a foundation for advancing future research on LALMs safety alignment by enabling the in-depth exposure of more powerful jailbreak threats, such as query-based audio editing, and by facilitating the development of effective defense mechanisms.
- Learning Anchor Planes for Classification26
The method needs only a few orthogonal anchor planes for coding, and it can linearize any (α, β, p)-Lipschitz smooth nonlinear function with a fixed expected value of the upper-bound approximation error on any high dimensional data.
- Fast Memory-Efficient Generalized Belief Propagation24
It is shown how the problem of subgraph matching can be formulated using this class of mrfs and thus, solved efficiently using the gbp approach and significantly outperform the state-of-the-art method.
- Near to Mid-term Risks and Opportunities of Open-Source Generative AI23
It is argued for the responsible open sourcing of generative AI models in the near and medium term and differential benefits and risks of open versus closed source AI and present potential risk mitigation, ranging from best practices to calls for technical and scientific contributions.
- Efficient minimization of higher order submodular functions using monotonic Boolean functions23
Novel mapping is defined that transform submodular functions of order $k$ into quadratic ones, which can be efficiently minimized in $O(n^3)$ time using a max-flow algorithm.
- Higher Order Potentials in End-to-End Trainable Conditional Random Fields23
Two types of higher order potentials can be included in a Conditional Random Field model embedded within a deep network to allow inference with the efficient and differentiable mean-field algorithm, making it possible to implement the CRF model as a stack of layers in adeep network.
- AlignGuard: Scalable Safety Alignment for Text-to-Image Generation22
AlignGuard is introduced, a method for safety alignment of T2I models that consistently outperforms the state-of-the-art on many benchmarks and establishes new practices for safety alignment in T2I networks.
- Zero-Shot Logit Adjustment22
This paper first analyzes of two properties of the generated pseudo unseen samples: bias and homogeneity, and performs variational Bayesian inference to back-derive the evaluation metrics, which reflects the balance of the seen and unseen classes.
- 22
- Object Proposals Estimation in Depth Image Using Compact 3D Shape Manifolds22
This work investigates the task of finding particular object shapes from a single depth image by exploiting the inherently low dimensionality in the object shape variations, which is discovered and encoded as a compact shape space.
- Learning to Hash Naturally Sorts21
This paper introduces Naturally-Sorted Hashing (NSH), a novel Sorted Noise-Contrastive Estimation (SortedNCE) loss that selectively picks positive and negative samples for contrastive learning, which allows NSH to mine data semantic relations during training in an unsupervised manner.
- 20
- PoseField: An Efficient Mean-Field Based Method for Joint Estimation of Human Pose, Segmentation, and Depth19
PoseField is proposed, a new highly efficient filter-based mean-field inference approach for jointly estimating human segmentation, pose, per-pixel body parts, and depth given stereo pairs of images.
- Improved Initialization and Gaussian Mixture Pairwise Terms for Dense Random Fields with Mean-field Inference19
A hierarchical mean-field approach where labelling from the coarser level is propagated to the finer level for better initialisation and a variation of Expectation Maximization (EM) for piecewise learning of the parameters of the mixture model determined by the maximum likelihood function is proposed.
- 19
- Neural Collapse Terminus: A Unified Solution for Class Incremental Learning and Its Variants18
A unified solution to the misalignment dilemma in the three tasks of CIL, LTCIL and FSCIL is offered andoretical analysis indicates that the method holds the neural collapse optimality in an incremental fashion regardless of data imbalance or data scarcity.
- How Robust is Unsupervised Representation Learning to Distribution Shift?18
This work develops a novel evaluation scheme with thelinear head trained on out-of-distribution (OOD) data, to isolate the performance of the pre-trained models from a potential bias of the linear head used for evaluation.
- 18
- A Generic CNN-CRF Model for Semantic Segmentation18
This work presents a CRF model were unary factors are dependent on a CNN, an efficient and scalable, maximum likelihood-based, learning procedure to infer all model parameters jointly and is the first to perform maximum likelihood learning for large-sized factor graphs with non-parametric potentials.
- Taxonomic Multi-class Prediction and Person Layout Using Efficient Structured Ranking18
An algorithm for structured output ranking that can be trained in a time linear in the number of samples under a mild assumption common to many computer vision problems: that the loss function can be discretized into a small number of values is proposed.
- 18
- 18
- Loong: Synthesize Long Chain-of-Thoughts at Scale through Verifiers17
The Loong Project is introduced: an open-source framework for scalable synthetic data generation and verification across a diverse range of reasoning-intensive domains and benchmarking LoongBench on a broad suite of both open-source and proprietary LLMs to evaluate domain coverage and reveal performance bottlenecks.
- A Semi-supervised Deep Generative Model for Human Body Analysis17
This work presents a deep generative model for human body analysis where the body pose and the visual appearance are disentangled in the latent space, allowing independent manipulation of pose and appearance and hence enables applications such as pose-transfer without being explicitly trained for such a task.
- OBJCUT for Face Detection17
This paper proposes a novel, simple and efficient method for face segmentation which works by coupling face detection and segmentation in a single framework using the OBJCUT formulation.
- Agent-X: Evaluating Deep Multimodal Reasoning in Vision-Centric Agentic Tasks16
Agent-X is introduced, a large-scale benchmark for evaluating vision-centric agents multi-step and deep reasoning capabilities in real-world, multimodal settings and proposes a fine-grained, step-level evaluation framework that assesses the correctness and logical coherence of each reasoning step and the effectiveness of tool usage throughout the task.
- 16
- 16
- Efficient Linear Programming for Dense CRFs16
This work develops a proximal minimization framework, where the dual of each proximal problem is optimized via block coordinate descent, and shows that each block of variables can be efficiently optimized.
- 16
- Flex3D: Feed-Forward 3D Generation with Flexible Reconstruction Model and Input View Curation15
Flex3D is a novel two-stage framework capable of leveraging an arbitrary number of high-quality input views, built upon a transformer architecture that can effectively process an arbitrary number of inputs, enabling efficient and detailed 3D generation.
- 15
- Building Models of Regular Scenes from Structure and Motion15
The method uses the camera models and point cloud typically generated by a structure-and-motion process as a starting point for developing a higher level model of the scene to improve the accuracy of fit.
- True Multimodal In-Context Learning Needs Attention to the Visual Context14
DARA is introduced, an efficient fine-tuning strategy that encourages models to attend to the visual context by rebalancing attention across visual and textual tokens and presents TrueMICL, an MICL-dedicated dataset that explicitly requires the integration of multimodal information-particularly visual content-for correct task completion.
- RanDumb: Random Representations Outperform Online Continually Learned Representations14
This study reveals the significant limitations of representation learning, particularly in low-exemplar and online continual learning scenarios, and challenges the prevailing assumptions about effective representation learning in online continual learning.
- Learning Visual Prompts for Guiding the Attention of Vision Transformers14
The proposed optimization-based visual prompting strategy is learned in a self-supervised manner without requiring annotations and without fine-tuning the vision transformer.
- 14
- MIP against Agent: Malicious Image Patches Hijacking Multimodal OS Agents13
It is shown that MIPs generalise across user prompts and screen configurations, and that they can hijack multiple OS agents even during the execution of benign instructions, which exposes critical security vulnerabilities in OS agents that have to be carefully addressed before their widespread deployment.
- Reliable Evaluation of Adversarial Transferability13
Contrary to the prevailing belief, the reevaluation revealed that the adversarial transferability across these diverse network types is notably diminished, and there is no single AE that can be transferred to all popular models.
- 13
- Mining Pixels: Weakly Supervised Semantic Segmentation Using Image Labels13
This work considers the task of learning a classifier for semantic segmentation using weak supervision, in this case, image labels specifying the objects within the image, and proposes minimizing the combination of the standard softmax loss and the KL divergence between the true latent posterior and the likelihood given by the CNN.
- Likelihood Models For Template Matching using the PDF Projection Theorem13
A novel method for estimating the likelihood PDFS accurately based on the PDF Projection Theorem is introduced, which provides the correct relation between the feature likelihood and the data likelihood, permitting the use of different types of features for different type of objects and still estimating consistent likelihoods.
- 13
- 12
- h1: Bootstrapping LLMs to Reason over Longer Horizons via Reinforcement Learning11
This work introduces a scalable method to bootstrap long-horizon reasoning capabilities using only existing, abundant short-horizon data, synthetically composes simple problems into complex, multi-step dependency chains of arbitrary length.
- Eigen-1: Adaptive Multi-Agent Refinement with Monitor-Based RAG for Scientific Reasoning11
A unified framework that combines implicit retrieval and structured collaboration and demonstrates how implicit augmentation and structured refinement overcome the inefficiencies of explicit tool use and uniform aggregation.
- Deconstructed Generation-Based Zero-Shot Model11
This paper aims to deconstruct the generator-classifier framework and provide guidance for its improvement and extension, and presents a simple method that outperforms SotAs on four public GZSL datasets, demonstrating the validity of the deconstruction.
- Learning Arbitrary Potentials in CRFs with Gradient Descent.11
A new inference and learning framework which can learn arbitrary pairwise CRF potentials is introduced based on the observation that CRF inference can be achieved via projected gradient descent and consequently, can be integrated in deep neural networks to allow for end-to-end training.
- Exploiting projective geometry for view-invariant monocular human motion analysis in man-made environments11
A methodology for view-invariant monocular human motion analysis in man-made environments in which some properties of projective geometry and the presence of numerous easy-to-detect straight lines are exploited.
- Simultaneous Human Segmentation, Depth and Pose Estimation via Dual Decomposition11
This work proposes a framework for estimating a detailed human skeleton in 3D from a stereo pair of images, and defines an energy function that incorporates the relationship between the segmentation results, the pose estimation results, and the disparity space image.
- Multiple-Instance Learning with Structured Bag Models10
Techniques for incorporating MIL constraints into Conditional Random Field models are introduced, thus providing a set of tools for constructing structured bag models, in which spatial dependencies are represented.
- Towards Interpretable Deep Local Learning with Successive Gradient Reconciliation9
This paper theoretically study non-greedy layer-wise training and shows that the convergence cannot be assured when the local gradient in a module w.r.t. its input is not reconciled with the local gradient in the previous module w.r.t. its output.
- 9
- Set-based Neural Network Encoding Without Weight Tying8
A neural network weight encoding method for network property prediction that utilizes set-to-set and set-to-vector functions to efficiently encode neural network parameters and introduces a pipeline to efficiently encode neural network layers that is adjustable to computational and memory constraints.
- 8
- 8
- 7
- 7
- A Robust Stereo Prior for Human Segmentation7
This work proposes a robust method of using a disparity space image to create a prior for human segmentation, and shows how the prior can be inserted into a dual decomposition formulation for stereo, segmentation and human pose estimation.
- 7
- 7
- Analyzing Convex Relaxations for MAP Estimation7
This chapter will focus on theoretical analysis for MAP estimation of discrete MRFs, with the aim of highlighting the importance of designing tight relaxations and the efficiency afforded by this approach.
- 7
- 7
- 7
- Dynamic Context-oriented Decomposition for Task-aware Low-rank Adaptation with Less Forgetting and Faster Convergence6
Context-oriented decomposition adaptation (CorDA), a novel method that initializes adapters in a task-aware manner, and develops CorDA++ in KPM, which not only achieves better fine-tuning performance than LoRA, but also mitigates the forgetting of pre-trained knowledge in both large language models and vision language models.
- Implicit neural representations for chemical reaction paths6
It is shown that neural networks can be optimized to represent minimum energy paths as continuous functions, offering a flexible alternative to discrete path-search methods such as Nudged Elastic Band and demonstrating promise for a universal reaction path representation.
- PSyDUCK: Training-Free Steganography for Latent Diffusion6
By overcoming the key challenges associated with latent diffusion model architectures, PSyDUCK sets a new standard for generative steganography, paving the way for scalable, real-world steganographic applications.
- 6
- 6
- 6
- 6
- 6
- Using strong shape priors for stereo6
This paper addresses the problem of obtaining an accurate 3D reconstruction from multiple views with a framework based on an object-specific Markov Random Field and shows how parameters of these models can be efficiently estimated by performing inference on the MRF using dynamic graph cuts.
- 5
- 5
- Body Language Based Individual Identification in Video Using Gait and Actions5
This report considers two different categories of individual recognition, first, individual recognition using "style of walk" i.e. gait and second " style of doing similar actions" in video sequences.
- Fitting multiple models to multiple images with minimal user interaction5
A method for modelling compound structures which intelligently divides this prior information into that which might be applied by the system and that which may not, in a manner which minimises the human input required.
- Rapid Interactive Modelling from Video with Graph Cuts5
The method is novel in that it uses information from several sources—video, sparse 3D points and user input—to fit models to a scene, and avoids common pitfalls of both automatic structure and motion algorithms, and image-based modelling pack ages.
- Lessons from reinforcement learning for biological representations of space4
This paper introduces a hand-crafted representation that has, by design, a high degree of geometric consistency and demonstrates that, in this case, information about the persistence of features as the camera translates can improve performance on the geometric tasks.
- 4
- 4
- StyP-Boost: A Bilinear Boosting Algorithm for Learning Style-Parameterized Classifiers4
A novel bilinear boosting algorithm is introduced, which extends the multi-class boosting framework of JointBoost to optimize a bilInear objective function, and style parameters are introduced to aid classification, where style is any factor which the classes vary with systematically, modeled by a vector quantity.
- Computer vision -- ECCV 2008 : 10th European Conference on Computer Vision, Marseille, France, October 12-18, 2008 : proceedings4
Image Segmentation in the Presence of Shadows and Highlights, and Key Object Driven Multi-category Object Recognition, Localization and Tracking and Stereo Matching.
- From Visual Query to Visual Portrayal4
This paper introduces a method to fuse the rendered novel views from all input images at each virtual view point by computing their intrinsic image and illuminations and demonstrates the performance of the prototype system on two landmark structures.
- An Object Category Specific mrf for Segmentation4
This chapter presents a principled Bayesian method for detecting and segmenting instances of a particular object category within an image, providing a coherent methodology for combining top down and bottom up cues.
- 4
- 3
- 3
- HelloFresh: LLM Evaluations on Streams of Real-World Human Editorial Actions across X Community Notes and Wikipedia edits3
HelloFresh is introduced, based on continuous streams of real-world data generated by intrinsically motivated human labelers, and it is found that HelloFresh yields a temporally consistent ranking.
- 3
- From Categories to Classifiers: Name-Only Continual Learning by Exploring the Web3
Overall, this paper underscores the potential of using uncurated webly-supervised data to mitigate the challenges associated with manual data labeling in continual learning.
- Certifying Ensembles: A General Certification Theory with S-Lipschitzness3
This work generalise Lipschitz continuity by introducing S-Lipschitzer classifiers, which are used to analyse the theoretical robustness of ensembles and results are precise conditions when ensemble of robust classifiers are more robust than any constituent classifier.
- 3
- 3
- 3
- Efficient Semidefinite Branch-and-Cut for MAP-MRF Inference3
Experiments show that the proposed algorithm achieves better approximation than the state-of-the-art methods within a variety of time budgets on challenging non-submodular MAP-MRF inference problems.
- 3
- Hand Tracking Using a Quadric Surface Model and Bayesian Filtering3
The effectiveness of the model-based 3D hand tracking technique is demonstrated by using it for tracking 3D articulated and non-rigid hand motion from monocular video sequences in front of a cluttered background.
- 2
- 2
- 2
- 2
- Spatio-Temporal Action Instance Segmentation and Localisation2
This work addresses the problem of action localisation and instance segmentation in which multiple concurrent actions of the same class may be segmented out of an image sequence by casting the action tube extraction as an energy maximisation problem in which configurations of region proposals in each frame are assigned a cost.
- GP-LVM for data consolidation2
A probabilistic interpretation of canonical correlation analysis with Gaussian process latent variable models for human pose estimation and model selection techniques for nonlinear dimensionality reduction.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- Faithful Knowledge Distillation1
A faithful imitation framework is introduced to discuss the relative calibration of confidences with regard to a teacher-student pair and the advantages of the increased verifiability of faithful distillation over alternative adversarial distillation methods are demonstrated.
- 1
- 1
- 1
- 1
- 1
- 1
- Template-Based Hand Detection and Tracking1
The effectiveness of the model-based 3D hand tracking technique is demonstrated by using it for tracking 3D articulated and non-rigid hand motion from monocular video sequences in front of a cluttered background.
- –
- GALILEO: Embodied AI scientist for autonomous therapeutic discovery in dynamic membrane systems–
These results establish retrieval-and-editing-based physical learning as a route for auditable, experimentally grounded interventions and transferable membrane-blocker rules.
- –
- –
- MedGenesis: Toward a World Model for Autonomous Clinical and Translational Research–
MedGenesis, a clinical artificial intelligence scientist built on a world-model reasoning loop that jointly updates a Latent Hypothesis Space and a Latent Action Space under expected information gain, uncertainty reduction, and a safety prior P(safe), and integrates longitudinal electronic health records (EHRs) via the Virtual Clinical Trajectory and Observation Representation (ViCTOR).
- ActCam: Zero-Shot Joint Camera and 3D Motion Control for Video Generation–
ActCam is presented, a zero-shot method for video generation that jointly transfers character motion from a driving video into a new scene and enables per-frame control of intrinsic and extrinsic camera parameters.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
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 2026-10-12. 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.