Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Richard I. HartleySuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works353 from public data
- Go-ICP: A Globally Optimal Solution to 3D ICP Point-Set Registration1,193
This paper presents the first globally optimal algorithm, named Go-ICP, for Euclidean (rigid) registration of two 3D point-sets under the inline-formula notation, and derives novel upper and lower bounds for the registration error function.
- The Visual Object Tracking VOT2016 Challenge Results776
The Visual Object Tracking challenge VOT2016 goes beyond its predecessors by introducing a new semi-automatic ground truth bounding box annotation methodology and extending the evaluation system with the no-reset experiment.
- Word-level Deep Sign Language Recognition from Video: A New Large-scale Dataset and Methods Comparison698
This paper introduces a new large-scale Word-Level American Sign Language (WLASL) video dataset, containing more than 2000 words performed by over 100 signers, and proposes a novel pose-based temporal graph convolution networks (Pose-TGCN) that model spatial and temporal dependencies in human pose trajectories simultaneously, which has further boosted the performance of the pose- based method.
- Deep Stacked Hierarchical Multi-Patch Network for Image Deblurring663
A deep {hierarchical multi-patch network} inspired by Spatial Pyramid Matching to deal with blurry images via a fine-to-coarse hierarchical representation and is the first real-time deep motion deblurring model for 720p images at 30fps.
- Rotation Averaging650
This paper is conceived as a tutorial on rotation averaging, summarizing the research that has been carried out in this area and contains many new results, which were developed to fill gaps in knowledge, answering fundamental questions such as radius of convergence of the algorithms, and existence of local minima.
- Learning Trajectory Dependencies for Human Motion Prediction584
A simple feed-forward deep network for motion prediction, which takes into account both temporal smoothness and spatial dependencies among human body joints, and design a new graph convolutional network to learn graph connectivity automatically.
- 517
- Deblurring by Realistic Blurring494
This paper proposes a new method which combines two GAN models, i.e., a learning-to-Blur GAN (BGAN) and learning- to-DeBlurGAN (DBGAN), in order to learn a better model for image deblurring by primarily learning how to blur images.
- Hierarchical Neural Architecture Search for Deep Stereo Matching438
This paper proposes the first end-to-end hierarchical NAS framework for deep stereo matching by incorporating task-specific human knowledge into the neural architecture search framework and optimizing the architectures of the entire pipeline jointly.
- Learning to Estimate Hidden Motions with Global Motion Aggregation436
This paper argues that the occlusion problem can be better solved in the two-frame case by modelling image self-similarities, and introduces a global motion aggregation module, a transformer-based approach to find long-range dependencies between pixels in the first image, and perform global aggregation on the corresponding motion features.
- Neural Aggregation Network for Video Face Recognition419
This NAN is trained with a standard classification or verification loss without any extra supervision signal, and it is found that it automatically learns to advocate high-quality face images while repelling low-quality ones such as blurred, occluded and improperly exposed faces.
- Deep Image Deblurring: A Survey415
This paper presents a comprehensive and timely survey of recently published deep-learning based image deblurring approaches, aiming to serve the community as a useful literature review.
- Lending Orientation to Neural Networks for Cross-View Geo-Localization406
A novel method which endows deep neural networks with the `commonsense' of orientation is proposed, given a ground-level spherical panoramic image as query input and a Siamese network which explicitly encodes the orientation of each pixel of the images.
- 373
- Spatial-Aware Feature Aggregation for Image based Cross-View Geo-Localization341
A new deep network is developed to explicitly address inherent differences between ground and aerial views, and introduces a feature aggregation strategy via learning multiple spatial embeddings to improve the robustness of feature representation.
- Kernel Methods on the Riemannian Manifold of Symmetric Positive Definite Matrices316
To encode the geometry of the manifold in the mapping, a family of provably positive definite kernels on the Riemannian manifold of SPD matrices is introduced, derived from the Gaussian kernel, but exploit different metrics on the manifold.
- MB-TaylorFormer: Multi-branch Efficient Transformer Expanded by Taylor Formula for Image Dehazing291
A new Transformer variant, named Multi-branch Transformer expanded by Taylor formula (MB-TaylorFormer), which can em-bed coarse to fine features more flexibly at the patch embedding stage and capture long-distance pixel interactions with limited computational cost.
- 282
- MB-TaylorFormer V2: Improved Multi-Branch Linear Transformer Expanded by Taylor Formula for Image Restoration277
The proposed model, named the second version of Taylor formula expansion-based Transformer (for short MB-TaylorFormer V2) has the capability to concurrently process coarse-to-fine features, capture long-distance pixel interactions with limited computational cost, and improve the approximation of the Taylor expansion remainder.
- Where Am I Looking At? Joint Location and Orientation Estimation by Cross-View Matching271
A Dynamic Similarity Matching network is designed to estimate cross-view orientation alignment during localization and improves state-of-the-art performance on large-scale geo-localization datasets.
- Kernel Methods on Riemannian Manifolds with Gaussian RBF Kernels270
It is shown that many popular algorithms designed for Euclidean spaces, such as support vector machines, discriminant analysis and principal component analysis can be generalized to Riemannian manifolds with the help of such positive definite Gaussian kernels.
- GridFormer: Residual Dense Transformer with Grid Structure for Image Restoration in Adverse Weather Conditions240
The GridFormer framework achieves state-of-the-art results on five diverse image restoration tasks in adverse weather conditions, including image deraining, dehazing, deraining & dehazing, desnowing, and multi-weather restoration.
- Align and Prompt: Video-and-Language Pre-training with Entity Prompts238
This paper proposes Align and Prompt: a new video-and-language pre-training framework (AlPro), which operates on sparsely-sampled video frames and achieves more effective cross-modal alignment without explicit object detectors.
- Optimal Feature Transport for Cross-View Image Geo-Localization234
A novel Cross-View Feature Transport (CVFT) technique to explicitly establish cross-view domain transfer that facilitates feature alignment between ground and aerial images, and is implemented as network layers, which transports features from one domain to the other, leading to more meaningful feature similarity comparison.
- UPnP: An Optimal O(n) Solution to the Absolute Pose Problem with Universal Applicability224
This paper presents the first PnP solution that unifies all the above desirable properties within a single algorithm, and compares its result to state-of-the-art minimal, non-minimal, central, and non-central PNP algorithms, and demonstrates universal applicability, competitive noise resilience, and superior computational efficiency.
- Few-Shot Action Recognition with Permutation-Invariant Attention218
This work builds on a C3D encoder for spatio-temporal video blocks to capture short-range action patterns and outperforms the state of the art on the HMDB51, UCF101, miniMIT datasets.
- 218
- Learning Image Matching by Simply Watching Video217
An unsupervised learning based approach to the ubiquitous computer vision problem of image matching that achieves surprising performance comparable to traditional empirically designed methods.
- Beyond Local Search: Tracking Objects Everywhere with Instance-Specific Proposals207
This work presents an object tracker that is not limited to a local search window and has ability to probe efficiently the entire frame and provides improved robustness for fast moving objects as well as for ultra lowframerate videos.
- Multiple views gait recognition using View Transformation Model based on optimized Gait Energy Image207
- NAF: Neural Attenuation Fields for Sparse-View CBCT Reconstruction197
A novel and fast self-supervised solution for sparse-view CBCT reconstruction (Cone Beam Computed Tomography) that requires no external training data and achieves state-of-the-art accuracy and spends reasonably short computation time.
- ApolloCar3D: A Large 3D Car Instance Understanding Benchmark for Autonomous Driving193
This paper contributes the first large scale database suitable for 3D car instance understanding – ApolloCar3D and builds various baseline algorithms with the state-of-the-art deep convolutional neural networks.
- MC-Blur: A Comprehensive Benchmark for Image Deblurring188
This paper constructs a new large-scale multi-cause image deblurring dataset (MC-Blur), including real-world and synthesized blurry images with different blur factors, and conducts extensive benchmarking studies to compare SOTA methods in different scenarios, analyze their efficiency, and investigate the buildataset’s capacity.
- 185
- 185
- The IKEA ASM Dataset: Understanding People Assembling Furniture through Actions, Objects and Pose182
IKEA ASM is introduced—a three million frame, multi-view, furniture assembly video dataset that includes depth, atomic actions, object segmentation, and human poses that enables the development of holistic methods, which integrate multi-modal and multi-View data to better perform on these tasks.
- Proposal-free Temporal Moment Localization of a Natural-Language Query in Video using Guided Attention175
Experimental results show the proposed dynamic filter-based guided attention mechanism outperforms state-of-the-art methods on these datasets, confirming the effectiveness of the method.
- HDR-NeRF: High Dynamic Range Neural Radiance Fields173
Experimental results on synthetic and real-world scenes validate that the High Dynamic Range Neural Radiance Fields (HDR-NeRF) can not only accurately control the exposures of synthesized views but also render views with a high dynamic range.
- 171
- 162
- Transferring Cross-Domain Knowledge for Video Sign Language Recognition158
A novel method is proposed that learns domain-invariant visual concepts and fertilizes WSLR models by transferring knowledge of subtitled news sign to them, and outperforms previous state-of-the-art methods significantly.
- 151
- Beyond Cross-view Image Retrieval: Highly Accurate Vehicle Localization Using Satellite Image146
This paper addresses the problem of vehicle-mounted camera localization by matching a ground-level image with an overhead-view satellite map with a differentiable Levenberg-Marquardt module that projects features from the satellite map to the ground-view, based on a relative camera pose.
- 145
- 144
- 127
- 126
- 124
- 117
- Underwater Image Restoration via Contrastive Learning and a Real-World Dataset111
This work constructed a large-scale real underwater image dataset, dubbed Heron Island Coral Reef Dataset (`HICRD’), for the purpose of benchmarking existing methods and supporting the development of new deep-learning based methods.
- Multi-level Motion Attention for Human Motion Prediction101
This work proposes to extract motion attention to capture the similarity between the current motion context and the historical motion sub-sequences to effectively exploit motion patterns from the long-term history to predict the future poses.
- Expanding the Family of Grassmannian Kernels: An Embedding Perspective101
This work introduces several positive definite Grassmannian kernels, including universal ones, and demonstrates their superiority over previously-known kernels in various tasks, such as classification, clustering, sparse coding and hashing.
- 99
- BlockFusion: Expandable 3D Scene Generation using Latent Tri-plane Extrapolation98
Experimental results indicate that BlockFusion is capable of generating diverse, geometrically consistent and unbounded large 3D scenes with unprecedented high-quality shapes in both indoor and outdoor scenarios.
- Channel Attention Based Iterative Residual Learning for Depth Map Super-Resolution95
This paper proposes a new framework for real-world DSR, which consists of four modules: an iterative residual learning module with deep supervision to learn effective high-frequency components of depth maps in a coarse-to-fine manner, and a depth refinement module to improve the depth map by TGV regularization and input loss.
- Canny-VO: Visual Odometry With RGB-D Cameras Based on Geometric 3-D–2-D Edge Alignment94
This paper reviews the classical problem of free-form curve registration and applies it to an efficient RGB-D visual odometry system called Canny-VO, as it efficiently tracks all Canny edge features extracted from the images.
- Accurate 3-DoF Camera Geo-Localization via Ground-to-Satellite Image Matching93
This work addresses the problem of ground-to-satellite image geo-localization by estimating the camera latitude, longitude and orientation by matching a query image captured at the ground level against a large-scale database with geotagged satellite images and proposes a new geometry-aware image retrieval pipeline to improve the coarse localization accuracy.
- ARVo: Learning All-Range Volumetric Correspondence for Video Deblurring92
Quantitative and qualitative experiments show that the proposed implicit method to learn spatial correspondence among blurry frames in the feature space performs favorably on both datasets against previous state-of-the-art methods, confirming the benefit of modeling all-range spatial correspondence for video deblurring.
- Learning Optical Flow from a Few Matches87
This paper proposes an alternative displacement representation, named Sparse Correlation Volume, which is constructed directly by computing the k closest matches in one feature map for each feature vector in the other feature map and stored in a sparse data structure.
- 86
- ConsistNet: Enforcing 3D Consistency for Multi-View Images Diffusion85
This paper proposes a novel method that can generate multiple images of the same object, as if they are captured from different viewpoints, while the 3D consistencies among those multiple generated images are effectively exploited.
- Rolling Shutter Camera Relative Pose: Generalized Epipolar Geometry84
The rolling shutter essential matrix is introduced, and its link to existing models such as the push-broom cameras is established, summarized in a clean hierarchy of multi-perspective cameras.
- 83
- Unsupervised Deep Epipolar Flow for Stationary or Dynamic Scenes83
Experimental results on various benchmarking datasets show that the Deep Epipolar Flow method achieves competitive performance compared with supervised methods and outperforms state-of-the-art unsupervised deep-learning methods.
- Open-World Stereo Video Matching with Deep RNN83
A novel Recurrent Neural Network (RNN) that takes a continuous stereo video as input, and directly predicts a depth-map at each frame without a pre-training process, and without the need of ground-truth depth-maps as supervision.
- A Simple Solution to the Six-Point Two-View Focal-Length Problem80
A 15th degree polynomial in the unknown focal-length estimation problem is derived based on the hidden-variable technique and will make a small but useful module in many computer vision systems.
- 79
- Beyond Monocular Deraining: Parallel Stereo Deraining Network Via Semantic Prior78
A Paired Rain Removal Network (PRRNet), which exploits both stereo images and semantic information is presented, and a Semantic-Aware Deraining Module (SADM) is developed which solves both tasks of semantic segmentation and deraining of scenes.
- Rethinking Class Relations: Absolute-relative Supervised and Unsupervised Few-shot Learning77
This work rethink the relations between class concepts, and proposes a novel Absolute-relative Learning paradigm to fully take advantage of label information to refine the image an relation representations in both supervised and unsupervised scenarios.
- 77
- Shape Interaction Matrix Revisited and Robustified: Efficient Subspace Clustering with Corrupted and Incomplete Data77
This paper revisits the Shape Interaction Matrix and reveals its connections to several recent subspace clustering methods, and derives a simple, yet effective algorithm to robustify the SIM and make it applicable to realistic scenarios where the data is corrupted by noise.
- Boosting 3-DoF Ground-to-Satellite Camera Localization Accuracy via Geometry-Guided Cross-View Transformer76
This paper designs a geometry-guided cross-view transformer that combines the benefits of conventional geometry and learnable cross-view transformers to map the ground-view observations to an overhead view and develops a neural pose optimizer with strong global information embedding ability to estimate the relative rotation between them.
- Monocular Dense 3D Reconstruction of a Complex Dynamic Scene from Two Perspective Frames76
This paper proposes a new approach for monocular dense 3D reconstruction of a complex dynamic scene from two perspective frames by applying superpixel over-segmentation to the image, and model a generically dynamic scene with a piecewise planar and rigid approximation.
- Ground-Plane-Based Absolute Scale Estimation for Monocular Visual Odometry74
This paper proposes a robust divide-and-conquer absolute scale estimation method based on the ground plane and camera height by analyzing the advantages and disadvantages of different approaches.
- Noise-Aware Unsupervised Deep Lidar-Stereo Fusion74
LidarStereoNet is presented, the first unsupervised Lidar-stereo fusion network, which can be trained in an end-to-end manner without the need of ground truth depth maps, and is proposed to incorporate the piecewise planar model into the network learning to further constrain depths to conform to the underlying 3D geometry.
- 74
- Geometry-Guided Street-View Panorama Synthesis From Satellite Imagery73
This paper develops a novel satellite to street-view projection (S2SP) module which learns the height map and projects the satellite image to the ground-level viewpoint, explicitly connecting corresponding pixels.
- 73
- Deep Two-View Structure-from-Motion Revisited69
This work proposes to revisit the problem of deep two-view SfM by leveraging the well-posedness of the classic pipeline by developing a scale-invariant depth estimation network that leverages epipolar geometry to reduce the search space, refine the dense correspondences, and estimate relative depth maps.
- 68
- The Thermal Infrared Visual Object Tracking VOT-TIR2016 Challenge Results68
A significant general improvement of results has been observed, which partly compensate for the more difficult sequences introduced into the dataset in VOT-TIR2016.
- 68
- Globally-Optimal Inlier Set Maximisation for Simultaneous Camera Pose and Feature Correspondence67
This work proposes a globally-optimal inlier set cardinality maximisation approach which jointly estimates optimal camera pose and optimal correspondences and shows that the method performs much more robustly than existing approaches, including on a large-scale outdoor data-set.
- Action Anticipation by Predicting Future Dynamic Images66
This paper presents a method for human action anticipation by predicting the most plausible future human motion using Dynamic Images and makes use of tailored loss functions to encourage a generative model to produce accurate future motion prediction.
- 66
- Scalable Dense Non-rigid Structure-from-Motion: A Grassmannian Perspective63
This paper proposes a new approach for dense NRSfM by modeling the problem on a Grassmann manifold that is scalable and more accurate than state-of-the-art methods, and more robust to noise and generalizes to highly nonlinear deformations.
- 62
- 62
- 62
- 61
- Cross-view and multi-view gait recognitions based on view transformation model using multi-layer perceptron60
A new multi-view gait recognition which estimates gait feature on one view using selected gait features from several other views using a well selected region of interest (ROI) on gaitfeature from another view is proposed.
- 59
- Spatio-temporal union of subspaces for multi-body non-rigid structure-from-motion59
This paper proposes a unified framework to jointly segment and reconstruct multiple non-rigid objects and proposes to exploit the structure of the scenes along both temporal direction and spatial direction, thus achieving a spatio-temporal representation.
- Robust Optical Flow Estimation of Double-Layer Images under Transparency or Reflection59
This paper proposes a robust solution to the problem of handling generic optical flow fields of two-frame images containing transparency or reflection by exploiting a generalized double-layer brightness consistency constraint connecting these two tasks, and utilizes the priors for both of them.
- Optimal Essential Matrix Estimation via Inlier-Set Maximization58
The globally optimal “rotation space search” method is extended to essential matrix estimation in the presence of feature mismatches or outliers, finding an explicit, geometrically meaningful essential manifold parametrization using a 5D direct product space of a solid 2D disk and a solid 3D ball.
- Dual Pixel Exploration: Simultaneous Depth Estimation and Image Restoration57
This paper proposes an end-to-end DDDNet (DP-based Depth and Deblur Network) to jointly estimate the depth and restore the image, and defines a re-blur loss, which reflects the relationship of the DP image formation process with depth information, to regularise the depth estimate in training.
- Multi-Level Second-Order Few-Shot Learning56
A Multi-level Second-order (MlSo) few-shots learning network for supervised or unsupervised few-shot image classification and few- shot action recognition is proposed, leveraging so-called power-normalized second-order base learner streams combined with features that express multiple levels of visual abstraction, and self-supervised discriminating mechanisms are used.
- 56
- 56
- Beyond Monocular Deraining: Stereo Image Deraining via Semantic Understanding55
A Paired Rain Removal Network (PRRNet) is presented, which exploits both stereo images and semantic information, and a Semantic-Aware Deraining Module (SADM) which solves both tasks of semantic segmentation and deraining of scenes, and a Semantic-Fusion Network (SFNet) and a View-Fusion Network (VFNet) which fuse semantic information and multi-view information respectively.
- 53
- 52
- Multi-Body Non-Rigid Structure-from-Motion52
This paper presents the first multi-body non-rigid structure-from-motion (SFM) method, which simultaneously reconstructs and segments multiple objects that are undergoing non- Rigid deformation over time.
- Bispectral analysis and recovery of images distorted by a moving water surface51
A new algorithm to recover a geometrically correct image of an object or scene from a set of images distorted by the wave motion of a water surface is proposed, based on higher-order spectra analysis-in particular, the bispectrum.
- Neural Reflectance for Shape Recovery with Shadow Handling49
This paper proposes a coordinate-based deep MLP (multilayer perceptron) to parameterize both the unknown 3D shape and the unknown reflectance at every surface point, able to leverage the observed photometric variance and shadows on the surface, and recover both surface shape and general non-Lambertian reflectance.
- 49
- 47
- Moving object detection and segmentation in urban environments from a moving platform47
An effective approach to detect and segment moving objects from two time-consecutive stereo frames, which leverages the uncertainties in camera motion estimation and in disparity computation and is robust against both global and local noise.
- 47
- Support Vector Machines and Their Application in Chemistry and Biotechnology47
Beneficial for chemical data analysis and the modeling of complex physic-chemical and biological systems, support vector machines show promise in a myriad of areas and this book enables non-mathematicians to understand the potential of SVMs and utilize them in a host of applications.
- 45
- The Alignment of the Spheres: Globally-Optimal Spherical Mixture Alignment for Camera Pose Estimation45
This work casts the problem as a 2D-3D mixture model alignment task and proposes the first globally-optimal solution to this formulation under the robust L2 distance between mixture distributions, guaranteeing global optimality without requiring a pose estimate.
- 45
- 45
- Robust Motion Segmentation with Unknown Correspondences43
This paper introduces an approach to performing motion segmentation without any prior knowledge of point correspondences in terms of Partial Permutation Matrices (PPMs) and aims to match feature descriptors while simultaneously encouraging point trajectories to satisfy subspace constraints.
- Superpixel Soup: Monocular Dense 3D Reconstruction of a Complex Dynamic Scene41
This work proposes a unified approach to solve the task of dense 3D reconstruction of a complex dynamic scene from images and shows that this approach provides an effective solution to the inherent scale-ambiguity in structure-from-motion under perspective projection.
- 41
- Optimizing over Radial Kernels on Compact Manifolds41
It is shown that positive definite radial kernels on the unit n-sphere, the Grassmann manifold and Kendall's shape manifold can be expressed in a simple form whose parameters can be automatically optimized within a support vector machine framework.
- LLDiffusion: Learning Degradation Representations in Diffusion Models for Low-Light Image Enhancement40
A degradation-aware learning scheme for LLIE using diffusion models, which effectively integrates degradation and image priors into the diffusion process, resulting in improved image enhancement.
- Structure from Recurrent Motion: From Rigidity to Recurrency38
This paper proposes a new method from a long monocular video sequence observing a non-rigid object performing recurrent and possibly repetitive dynamic action to reduce NRSfM problems to rigid ones provided that certain recurrency condition is satisfied.
- 38
- Interacting Hand-Object Pose Estimation via Dense Mutual Attention37
This work proposes a novel dense mutual attention mechanism that is able to model fine-grained dependencies between the hand and the object and outperforms state-of-the-art methods.
- 36
- Divide and Conquer: Efficient Density-Based Tracking of 3D Sensors in Manhattan Worlds36
This paper proposes a divide-and-conquer technique during which the estimation of the rotation and the three degrees of freedom of the translation are all decoupled from one another, and pushes the idea towards a highly efficient and reliable solution for full 6DoF motion estimation with only depth information.
- A Framework for Shape Analysis via Hilbert Space Embedding36
This work proposes to map shapes on Kendall's shape manifold to a high dimensional Hilbert space where Euclidean geometry applies, and introduces a kernel on this manifold that permits such a mapping, and proves its positive definiteness.
- 36
- Improving GAN Equilibrium by Raising Spatial Awareness35
This work aims to alleviate the issue of D dominating the competition in GANs, and proposes to align the spatial awareness of G with the attention map induced from D, effectively lessen the information gap between D and G.
- 33
- 33
- Element-Wise Factorization for N-View Projective Reconstruction33
This paper recasts the problem as matrix factorization as element-wise factorization, leading to a convenient and efficient semi-definite program formulation, and presents an optimal solution to projective factorization for structure and motion, based on the same principle of low-rank factorization.
- 32
- 32
- Frankenstein: Generating Semantic-Compositional 3D Scenes in One Tri-Plane31
Frankenstein, a diffusion-based framework that can generate semantic-compositional 3D scenes in a single pass, demonstrates promising results in generating room interiors as well as human avatars with automatically separated parts.
- MAVIS: Multi-Camera Augmented Visual-Inertial SLAM using SE2(3) Based Exact IMU Pre-integration31
An improved IMU pre-integration formulation based on the exponential function of an automorphism of SE2(3), which can effectively enhance tracking performance under fast rotational motion and extended integration time is introduced.
- RGB-based Category-level Object Pose Estimation via Decoupled Metric Scale Recovery31
This paper proposes a novel pipeline that decouples the 6D pose and size estimation to mitigate the influence of imperfect scales on rigid transformations, and advocates using the RANSAC-PnP algorithm to robustly solve for 6D object pose.
- Event-guided Multi-patch Network with Self-supervision for Non-uniform Motion Deblurring30
A novel self-supervised event-guided deep hierarchical Multi-patch Network (MPN) to deal with blurry images and videos via fine-to-coarse hierarchical localized representations and achieves the state of the art on the GoPro and VideoDeblur datasets with a 40% faster runtime compared to current multi-scale methods.
- 30
- 29
- Lighting, Reflectance and Geometry Estimation from 360° Panoramic Stereo29
Both quantitative and qualitative experiments show that the method, benefiting from the 360° observation of the scene, outperforms prior state-of-the-art methods and enables more augmented reality applications such as mirror-objects insertion.
- Multi-view 3D Reconstruction of a Texture-less Smooth Surface of Unknown Generic Reflectance29
This paper proposes a simple and practical solution to overcome the challenge of establishing cross-view correspondences where photometric constancy is violated based on a co-located camera-light scanner device and develops an optimization algorithm that robustly recovers globally optimal shape and reflectance even from a random initialization.
- 29
- Supervised dimensionality reduction via sequential semidefinite programming29
This work directly optimize the trace quotient as a quasi-linear semidefinite optimization problem, which can be solved globally and efficiently using standard off-the-shelf semideFinite programming solvers and allows one to enforce additional constraints on the projection matrix.
- WildLight: In-the-wild Inverse Rendering with a Flashlight28
This paper proposes a practical photometric solution for the challenging problem of in-the-wild inverse rendering under unknown ambient lighting using smartphone's built-in flashlight to decompose image intensities into two photometric components – a static appearance corresponds to ambient flux, plus a dynamic reflection induced by the moving flashlight.
- 28
- Deep Novel View Synthesis from Colored 3D Point Clouds27
A new deep neural network is proposed which takes a colored 3D point cloud of a scene as input, and synthesizes a photo-realistic image from a novel viewpoint and conducts numerous experiments on public datasets to validate the method in terms of quality of the synthesized views.
- 27
- Privacy Assessment on Reconstructed Images: Are Existing Evaluation Metrics Faithful to Human Perception?26
A learning-based measure called SemSim is proposed to evaluate the Semantic Similarity between the original and reconstructed images and it is shown that SemSim has a significantly higher correlation with human judgment compared with existing metrics.
- 26
- MEGANE: Morphable Eyeglass and Avatar Network25
This work proposes a 3D compositional morphable model of eyeglasses that accurately incorporates high-fidelity geometric and photometric interaction effects and models global light transport effects, such as casting shadows between faces and glasses.
- Rolling Shutter Inversion: Bring Rolling Shutter Images to High Framerate Global Shutter Video25
This paper represents the very first rolling-shuttered temporal super-resolution deep-network that is able to recover high framerate global-shutter videos from just two RS frames, and exploits the underlying spatial-temporal geometric relationships within a deep learning framework.
- A Convex Programming Approach to the Trace Quotient Problem25
This paper reformulate the original non-convex problem such that it can be solved by efficiently solving a sequence of semidefinite feasibility problems and is therefore globally optimal.
- Where’s the Weet-Bix?25
Through the study this new retrieval problem proves itself to be a challenging task and an instant application of it is to help the customer find what they want without physically wandering around the shelves but a wide range of potential applications could be expected.
- Every Moment Matters: Detail-Aware Networks to Bring a Blurry Image Alive24
A detail-aware network with three consecutive stages to improve the reconstruction quality by addressing specific aspects in the recovery process and produces sharp image sequences with rich details and subtle motion, outperforming the state-of-the-art methods.
- Cousin Network Guided Sketch Recognition via Latent Attribute Warehouse24
This paper proposes to transfer the knowledge of a network learned from natural images to a sketch network - a new deep net architecture which is term as cousin network, which guides a sketch-recognition network to extract more relevant features that are close to those of natural images, via adversarial training.
- 24
- 24
- 23
- “Maximizing Rigidity” Revisited: A Convex Programming Approach for Generic 3D Shape Reconstruction from Multiple Perspective Views23
This work revisits the principle of “maximizing rigidity” in structure-from-motion literature, and develops a unified theory which is applicable to both rigid and non-rigid structure reconstruction in a rigidity-agnostic way.
- Seeing Through the Glass: Neural 3D Reconstruction of Object Inside a Transparent Container22
This paper uses an existing neural reconstruction method (NeuS) that implicitly represents the geometry and appearance of the inner subspace of the scene as two distinct sub-spaces, inside and outside the transparent enclosure, and develops a hybrid rendering strategy that combines volume rendering with ray tracing.
- Self-Supervised Visibility Learning for Novel View Synthesis22
This paper proposes an end-to-end NVS framework and designs a source-view visibility estimation (SVE) module to determine the visibility of the target-view voxels in each source view, thus significantly alleviating error accumulation in view synthesis.
- 22
- 22
- StraightPCF: Straight Point Cloud Filtering21
This paper introduces StraightPCF, a new deep learning based method for point cloud filtering that works by moving noisy points along straight paths, thus reducing discretization errors while ensuring faster convergence to the clean surfaces.
- View Consistent Purification for Accurate Cross-View Localization21
This paper proposes a fine-grained self-localization method for outdoor robotics that utilizes a flexible number of onboard cameras and readily accessible satellite images, and is the first sparse visual-only method that enhances perception in dynamic environments by detecting view-consistent key points and their corresponding deep features from ground and satellite views.
- End-to-end Learning for Inter-Vehicle Distance and Relative Velocity Estimation in ADAS with a Monocular Camera20
A monocular camera based inter-vehicle distance and relative velocity estimation method based on end-to-end training of a deep neural network that integrates multiple visual clues provided by any two time-consecutive monocular frames, which include deep feature clue, scene geometry clue, as well as temporal optical flow clue.
- 20
- 20
- Plane-Based Calibration and Auto-calibration of a Fish-Eye Camera20
By taking images of a planar scene with a fish-eye camera, this method automatically estimates the centre-of-distortion (COD), distortion parameters, and other conventional camera intrinsics by a three-stage algorithm.
- Stereo Matching in Time: 100+ FPS Video Stereo Matching for Extended Reality19
A novel video stereo synthetic dataset that comprises photorealistic renderings of various indoor scenes and realistic camera motion captured by a moving VR/AR head-mounted display (HMD) is introduced and demonstrated that it achieves superior performance compared to the current state-of-the-art, making it a strong contender for real-time stereo matching in VR/AR applications.
- 19
- Stereo Computation for a Single Mixture Image19
This work gives a novel deep-learning based solution to the original problem of stereo computation from a single mixture image, by jointly solving the two subtasks of image layer separation as well as stereo matching.
- 19
- GUS-IR: Gaussian Splatting With Unified Shading for Inverse Rendering18
This paper starts by analyzing and comparing two prominent shading techniques popularly used for inverse rendering, forward shading and deferred shading, effectiveness in handling complex materials and proposes a unified shading solution that combines the advantages of both techniques for better decomposition.
- GS-2DGS: Geometrically Supervised 2DGS for Reflective Object Reconstruction18
This work proposes a novel reconstruction method called GS-2DGS for reflective objects based on 2D Gaussian Splatting (2DGS), which combines the rapid rendering capabilities of Gaussian Splatting with additional geometric information from foundation models and achieves performance comparable to SDF-based methods while being an order of magnitude faster.
- 18
- DeepSimHO: Stable Pose Estimation for Hand-Object Interaction via Physics Simulation18
This paper presents DeepSimHO: a novel deep-learning pipeline that combines forward physics simulation and backward gradient approximation with a neural network that noticeably improves the stability of the estimation and achieves superior efficiency over test-time optimization.
- Deep Non-Rigid Structure-From-Motion: A Sequence-to-Sequence Translation Perspective17
The Context Layer is proposed, which enables the deep learning framework to effectively impose overall constraints on sequences based on the structural characteristics of non-rigid sequences based on the structural characteristics of non-rigid sequences.
- 3D Scene Reconstruction with an Un-calibrated Light Field Camera17
A set of new equations are established and solved for self-calibration and 3D metric reconstruction specifically designed for a light field camera and a new concept of “rays of the absolute conic (RAC)” defined as a special quadric in 5D projective space is proposed.
- 6DoF Object Pose Estimation via Differentiable Proxy Voting Regularizer17
This paper develops a simple yet effective differentiable proxy voting regularizer (DPVR) which mimics the hypothesis selection in the voting procedure and is able to train the network in an end-to-end manner.
- Robust Multi-Body Feature Tracker: A Segmentation-Free Approach16
A segmentationfree solution to multi-body feature tracking that bypasses the motion assignment step and reduces to solving a series of subproblems with closed-form solutions is introduced.
- Semi-dense visual odometry for RGB-D cameras using approximate nearest neighbour fields15
A robust and efficient semidense visual odometry solution for RGB-D cameras that predominantly outperforms existing state-of-the-art methods is presented.
- 15
- Training a neural network for moment based image edge detection15
Tests showed the proposed neural network edge detector is superior to traditional operators and functions as a neural pattern classifier that directly estimates the posterior probability from the training data set.
- Inverting the Imaging Process by Learning an Implicit Camera Model14
This paper proposes a novel implicit camera model which represents the physical imaging process of a camera as a deep neural network, and devise an implicit blur generator and an implicit tone mapper to model the aperture and exposure of the camera's imaging process.
- 14
- 14
- 14
- 14
- 3D Geometry-Aware Semantic Labeling of Outdoor Street Scenes14
This paper proposes a 3D semantic labeling method to label outdoor street scenes whenever a dense depth map is available, and demonstrates that directly performing 3D convolution in the framework of a residual connected 3D voxel top-down modulation network can lead to superior results.
- 14
- 14
- Interactive color image segmentation with linear programming14
A linear programming (LP) approach for segmenting a color image into multiple regions, which has a simpler mathematical formulation, and a far lower computational complexity than the recently proposed semi-definite programming (SDP)-based approach.
- Conformal spherical representation of 3D genus-zero meshes14
This paper describes an approach of representing 3D shape by using a set of invariant spherical harmonic coefficients after conformal mapping, where the modification is based on Mobius factorization and aims at obtaining a canonical conformal mapping.
- Prompting Future Driven Diffusion Model for Hand Motion Prediction13
A prompt-based Future Driven Diffusion Model (PromptFDDM) for predicting hand motion with guidance and prompts is proposed, which achieves state-of-the-art performance in both first-and third-person perspectives.
- Homography Guided Temporal Fusion for Road Line and Marking Segmentation13
A Homography Guided Fusion module to exploit temporally-adjacent video frames for complementary cues facilitating the correct classification of the partially occluded road lines or markings is proposed and shown to lead to a light-weight network with significantly improved performances in speed and accuracy.
- 13
- Automatic Gloss Dictionary for Sign Language Learners13
GlossFinder can lower the barrier in sign language learning by addressing the common problem of sign finding and make it accessible to the wider community, according to a user study conducted among sign language speakers of varying ability.
- CropMix: Sampling a Rich Input Distribution via Multi-Scale Cropping13
This work demonstrates that CropMix can be seamlessly applied to virtually any training recipe and neural network architecture performing classification tasks, and is of benefit to both contrastive learning and masked image modeling towards more powerful representations.
- Transcribing Natural Languages for the Deaf via Neural Editing Programs13
This work designs a new neural agent that learns to synthesize and execute editing programs, conditioned on sentence contexts and partial editing results, that outperforms previous glossification models by a large margin.
- 13
- 13
- Low-Rank Kernel Subspace Clustering.13
This paper proposes to learn a low-rank kernel mapping, with which the mapped data in feature space are not only low- rank but also self-expressive, such that the low-dimensional subspace structures are present and manifested in the high-dimensional feature space.
- Gaussian process regression-based robust free space detection for autonomous vehicle by 3-D point cloud and 2-D appearance information fusion13
A stereo vision-based robust free space detection method is proposed which mainly depends on geometry information and Gaussian process regression and Bayesian framework and conditional random field inference are applied to fuse the multimodal information including 2-D image and 3-D point geometric information.
- 13
- Probability Density Geodesics in Image Diffusion Latent Space12
This work shows that geodesics can be computed in diffusion latent space, where the norm induced by the spatially-varying inner product is inversely proportional to the probability density.
- Ranking Models in Unlabeled New Environments12
The problem of ranking models in unlabeled new environments is introduced, a proxy dataset that is fully labeled and well reflects the true model rankings in a given target environment is proposed, and a carefully constructed proxy set effectively captures relative performance ranking in new environments.
- Deep robust image deblurring via blur distilling and information comparison in latent space12
A latent contrastive model, Blur Distilling and Information Reconstruction Networks (BDIRNet), to learn image prior and improve the robustness of deep deblurring and achieve improved and robust results compared to recent state-of-the-art methods.
- Full View Optical Flow Estimation Leveraged From Light Field Superpixel12
This paper presents a full view optical flow estimation method for plenoptic imaging that employs the structure delivered by the four-dimensional light field over multiple views making use of superpixels to recover the optical flow and depth maps.
- 12
- A learning-based markerless approach for full-body kinematics estimation in-natura from a single image12
The method opens new vistas to robust user-interaction free estimation of full body kinematics, a prerequisite to motion analysis, and achieves a score of 91.1 and 69.1 on mean Probability of Correct Keypoint measure and Average Precision of Keypoints measure for the frontal and sagittal datasets respectively.
- Learning Camera Pose from Optical Colonoscopy Frames Through Deep Convolutional Neural Network (CNN)12
It is shown that transfer learning from surgical simulation to actual endoscopic based surgery is a possible approach for deep learning technologies and could be estimated with higher accuracy and speed than feature based computer vision methods such as the classical structure from motion pipeline.
- 11
- Unsupervised 3D Pose Estimation with Non-Rigid Structure-from-Motion Modeling11
A new modeling method for human pose deformations and an accompanying diffusion-based motion prior is proposed and evaluated on mainstream datasets and obtained superior results outperforming the state-of-the-art.
- 11
- 11
- 11
- 11
- Camera lens radial distortion correction using two-view projective invariants11
A system of algebraic equations is derived that relates the invariants to the distortion parameters to be found and a noniterative procedure to solve the equations system, and a kernel-voting scheme to select the best root.
- 11
- Geometry-Guided Cross-View Diffusion for One-to-Many Cross-View Image Synthesis10
A Geometry-guided Cross-view Condition (GCC) strategy is introduced to establish explicit geometric correspondences between satellite and street-view features, which enables the geometry ambiguity introduced by camera pose between image pairs to be resolved, boosting the performance of cross-view image synthesis.
- PlückerNet: Learn to Register 3D Line Reconstructions¨10
Experiments on both indoor and outdoor datasets show that registration (rotation and translation) precision of the neural network based method outperforms baselines significantly.
- 10
- 10
- 10
- 10
- 10
- 10
- 10
- LAM3D: Large Image-Point-Cloud Alignment Model for 3D Reconstruction from Single Image9
A novel framework, the Large Image and Point Cloud Alignment Model (LAM3D), which utilizes 3D point cloud data to enhance the fidelity of generated 3D meshes and achieves state-of-the-art high-fidelity 3D mesh reconstruction from a single image in just 6 seconds.
- 9
- CDA: A Contrastive Data Augmentation Method for Alzheimer’s Disease Detection9
A novel contrastive data augmentation method, which simulates the cognitive impairment of a patient by randomly deleting a proportion of text from the transcript to create negative samples, which achieves the best performance among language-based models.
- Diffeomorphic Neural Surface Parameterization for 3D and Reflectance Acquisition9
Key to the method is a velocity field shape representation that drives the canonical surface to target shape through time, which yields an exceptionally clean formulation that can be optimized by standard gradient descent without initialization, and works with both near-field and distant light source.
- Displacement-Invariant Cost Computation for Stereo Matching9
This work proposes a displacement-invariant cost computation module to compute the matching costs without needing a 4D feature volume, and proposes an entropy-based refinement strategy to refine the computed disparity map, which further improves the speed by avoiding the need to compute a second disparity map on the right image.
- 9
- 9
- 9
- 9
- 9
- 8
- 8
- 8
- 8
- 8
- 8
- 7
- 7
- 7
- 7
- Model-Free Multiple Object Tracking with Shared Proposals7
A novel model-free framework that incorporates generic category-independent object proposals without the need to pretrain any object detectors is presented, which significantly improves the search efficiency in comparison to the traditional dense sampling approach.
- 7
- 7
- Inverse tensor transfer with applications to novel view synthesis and multi-baseline stereo7
A new tensor transfer-based novel view synthesis (NVS) method that operates inversely in the sense that it transfers a pixel from the novel view image back to the real images, alleviating many common problems with conventional methods.
- 7
- 6
- 6
- Uncertainty Estimation of Dense Optical Flow for Robust Visual Navigation6
A novel dense optical-flow algorithm to solve the monocular simultaneous localisation and mapping (SLAM) problem for ground or aerial robots and demonstrates enhanced robustness and accuracy for the public autonomous car dataset (KITTI) and aerial monocular dataset.
- 6
- Deep Stochastic Attraction and Repulsion Embedding for Image Based Localization.6
This paper represents a place as a set of exemplar images depicting the same landmarks, instead of some pre-defined geographic locations by partitioning the world, and proposes a new Stochastic Attraction and Repulsion Embedding loss function to facilitate the competitive learning.
- Automated detection and tracking of slalom paddlers from broadcast image sequences using cascade classifiers and discriminative correlation filters6
A Periodically Prior Regularised Discriminative Correlation Filters (PPRDCF) framework is proposed, which exploits recent successful Discrim inative Correlations Filters with a periodic regularisation by a prior that constitutes a rich discriminative cascade classifier.
- 5
- 5
- Invariants for Discrete Structures – An Extension of Haar Integrals over Transformation Groups to Dirac Delta Functions5
This paper extends the methodology to derive invariants from discrete structures, like polygons or 3D-meshes by summing over monomials of discrete features of local support by extending Haar integrals over the Euclidean transformation group to Dirac delta functions.
- JADE: Joint-Aware Latent Diffusion for 3D Human Generative Modeling4
JADE is introduced, a generative framework that learns the variations of human shapes with fined-grained control that decomposes human bodies into skeleton structures, modeled by joint positions, and local surface geometries, characterized by features attached to each joint.
- 4
- Distance Based Image Classification: A solution to generative classification’s conundrum?4
This work proposes a new generative model in which semantic factors are accommodated by shell theory’s Wen-Yan et al. hierarchical generative process and non-semantic factors by an instance specific noise term, and uses the model to develop a classification scheme which suppresses the impact of noise while preserving semantic cues.
- 4
- 4
- 4
- A Motion Free Approach to Dense Depth Estimation in Complex Dynamic Scene.4
This work shows that, given per-pixel optical flow correspondences between two consecutive frames and the sparse depth prior for the reference frame, it can recover the dense depth map for the successive frames without solving for motion parameters.
- 4
- Compressive Evaluation in Human Motion Tracking4
This work proposes a compressive anneal particle filter to exploit sparsity existing in image-based human motion tracking, and introduces a progressive multilevel wavelet decomposition staged at eachAnneal layer to accelerate the compressive evaluation in a coarse-to-fine fashion.
- 4
- New 3D Fourier Descriptors for Genus-Zero Mesh Objects4
A newly developed surface parametrization method–the discrete conformal mapping (DCM)—is used to propose a 3D Fourier Descriptor (3D-FD) for representing and recognizing arbitrarily-complex genus-zero mesh objects.
- A simple rectification method for linear multi-baseline stereovision system4
Instead of using the general 8-parameter homography transform, a two-step, virtual rotation method is applied for rectification, which results in a more specific transform that has only 3 parameters, and more stability.
- Physically Plausible Color Correction for Neural Radiance Fields3
A novel color correction module that simulates the physical color processing in cameras to be embedded in NeRF, enabling the unified color NeRF reconstruction and demonstrating that the method is superior to baseline methods in both quantitative and qualitative evaluations of color correction and color consistency.
- 3
- Spatial Steerability of GANs via Self-Supervision from Discriminator3
This work designs randomly sampled Gaussian heatmaps to be encoded into the intermediate layers of generative models as spatial inductive bias, and incorporates DragGAN into the framework, which facilitates fine-grained manipulation within a reasonable time and supports a coarse-to-fine editing process.
- 3
- Deep Video Restoration for Under-Display Camera3
This work proposes a novel transformer-based baseline method that adaptively enhances degraded videos of UDC-VR and builds the first large-scale UDC video restoration dataset called PexelsUDC, which includes two subsets corresponding to different displays for UDC.
- 3
- Few-shot Action Recognition via Improved Attention with Self-supervision3
This paper tackles a more challenging task of few-shot action-recognition from video clips with a simple framework which is both flexible and easy to implement and compares favorably with several state-of-the-art baselines on HMDB51, miniMIT and UCF101 datasets.
- 3
- 3
- 3
- 3
- 3
- 2
- Fast-dLLM++: Fréchet Profile Decoding for Faster Diffusion LLM Inference2
Experiments show that the theoretical improvement translates directly into empirical gains: profile-aware selection improves the accuracy--throughput frontier by exploiting safe parallelism that weakest-token rules miss, achieving up to 37% higher throughput at comparable accuracy.
- Mitigating Multimodal Hallucinations via Gradient-based Self-Reflection2
Gradient-based Influence-Aware Constrained Decoding (GACD), an inference-based method, that addresses both biases without auxiliary models, and is readily applicable to existing models without finetuning, is proposed.
- 2
- 2
- Learning To Segment Dominant Object Motion From Watching Videos2
Inspired by a layered image representation, a technique to group pixel regions according to their affine parametric motion is introduced, which enables the network to learn segmentation of the dominant foreground object using only RGB image pairs as input for both training and inference.
- 2
- Rethinking Class Relations: Absolute-relative Few-shot Learning.2
A novel absolute-relative learning paradigm is proposed to fully take advantage of label information to refine the image representations and correct the relation understanding and improves the performance of several the state-of-the-art models on publicly available datasets.
- Estimating the projected frontal surface area of cyclists from images using a variational framework and statistical shape and appearance models2
The method successfully estimates pFSA of cyclists and opens new vistas for exploration of the relationship with aerodynamic drag, using a statistical cyclist shape and appearance models as priors to encourage the evolving curve to arrive at the desired solution.
- 2
- Gradual Sampling and Mutual Information Maximisation for Markerless Motion Capture2
Noticing the limitation of silhouettes in resolving gesture ambiguity, this work incorporates appearance information in an illumination invariant way by maximising Mutual Information between an appearance model and the observation, which strengthens the effectiveness of the better-behaved observation likelihood.
- 2
- 2
- Feature Matching and Pose Estimation Using Newton Iteration2
This paper proposes a new method that is able to estimate the transformation and find the correct matches simultaneously, and formulate the problem as a nearest-matrix problem, and provides a different numerical technique.
- Rectification-Free Multibaseline Stereo for Non-ideal Configurations2
A new algorithm to process non-ideally configured multibaseline stereo system, which not only avoids the rectification procedure but also remains the efficiency of SSSD at the same time.
- Discrete Conformal Shape Representation and Reconstruction of 3D Mesh Objects2
A conformal representation for genus-zero mesh objects is proposed, by using the discrete conformal mapping technique and a new method to reconstruct the original shape from its conformal representations is proposed.
- 2
- 1
- Level-Set Parameters: Novel Representation for 3D Shape Analysis1
The promise of the novel representations of level-set parameters from signed distance functions as a novel, continuous, and numerical representation of 3D shapes is demonstrated through applications in shape classification, retrieval, and 6D object pose estimation.
- 1
- 1
- Unsupervised Video Interpolation by Learning Multilayered 2.5D Motion Fields1
A self-supervised approach to video frame interpolation that requires only a single video posed as a set of layers, which learns the video as a space-time continuum, allowingframe interpolation at any temporal resolution.
- Benchmarking Deep Deblurring Algorithms: A Large-Scale Multi-Cause Dataset and A New Baseline Model.1
- 1
- 1
- Non-Linear Subspace Clustering with Learned Low-Rank Kernels1
This paper proposes to learn a low-rank kernel matrix, with which mapped data in feature space are not only low- rank but also self-expressive, and the low-dimensional subspace structures of the (implicitly) mapped data are retained and manifested in the high-dimensional feature space.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- Auto-Calibration of a Compound-Type Omnidirectional Camera1
An algorithm for automatically removing lens distortion by exploiting planar projective invariants is capable of estimating centre-of-distortion and distortion parameters efficiently without using any nonlinear iteration.
- –
- TMEDRP: decoding tumor-intrinsic and microenvironmental signatures for clinical drug response prediction–
TMEDRP is a novel two-stage learning framework that integrates tumor-intrinsic signatures and TME-specific components for enhanced clinical drug response prediction and exhibits promising zero-shot generalization across unseen cancer lineages and novel compounds, suggesting its robustness across diverse scenarios.
- –
- Mutual learning for joint disease detection and severity prediction reveals multimodal pathogenesis for neurodegenerative disorders–
Without bells and whistles, Pa-MACRO establishes new state-of-the-art results in various settings while maintaining superior interpretability, verifying its power and versatility in revealing the pathogenesis of neurodegenerative disorders.
- –
- –
- –
- –
- Neural Plenoptic Sampling: Learning Light-Field from Thousands of Imaginary Eyes–
This paper adopts a simple Multi-Layer Per-ceptron (MLP) network as a universal function approximator to learn the plenoptic function at every position in the space of interest and trains the MLP to regress the plenoptic function for the space.
- Faster Stochastic Variance Reduction Methods for Compositional MiniMax Optimization–
This work introduces ADAptive NSTORM (ADA-NSTORM) with adaptive learning rates, proving it achieves the same sample complexity while experiments demonstrate greater effectiveness and significantly advances compositional minimax optimization.
- Stereo Unstructured Magnification: Multiple Homography Image for View Synthesis–
A novel multiple homography image (MHI) representation, comprising of a set of scene planes with fixed normals and distances, is proposed, which achieves superior performance for view synthesis qualitatively and quantitatively, especially for cases when the cameras undergo rotations.
- –
- 3D Human Pose Estimation with 2D Human Pose and Depthmap–
A new 3D human pose estimation model that takes an estimated 2D pose and the depthmap of the2D pose as input to estimate 3DHuman3.6m dataset and can achieve better performance than the SLM, and that the model can align the 2D landmark data with the depth map automatically.
- –
- –
- –
- Anti-occlusion Light-Field Optical Flow Estimation Using Light-Field Super-Pixels–
A novel method for anti-occlusion optical flow estimation in a dynamic light field by model the light field superpixel (LFSP) as a slanted plane in 3D and optimizing the motion of the occluded pixels in central view slice can be optimized by the un-occluded pixels in other views.
- –
- –
- –
- –
- Reliable Point Correspondences in Scenes Dominated by Highly Reflective and Largely Homogeneous Surfaces–
A novel method that uses descriptors evaluated along points in image edges to obtain a sufficiently high proportion of correct point correspondences is proposed that gives better results in recovering the epipolar geometry in scenes dominated by highly reflective and homogeneous surfaces.
- –
- –
- An Effective M-ary Spread Spectrum Communication Scheme–
Computer simulation show the validity of the proposed scheme under additional white gauss noise channel models and it is effective in improving the bandwidth efficiency and anti-intercept capability.
- –
- –
- –
- –
- Ring-like Buffering Technology in Data Acquisition Application–
Analyzing the insufficiency of the double buffering technology, this paper discusses how to realize continual and infinite-length data acquisition with ring-like buffer technology, and describes its work mechanism and realization method emphatically.
- –
- –
- Fast algorithm for DCT domain image up-conversion–
This paper presents a fast D CT domain image up-conversion algorithm, and a DCT domain linear filter, which only involves matrix manipulation.
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.