Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks112 from public data
- Residual Attention Network for Image Classification3,863
The proposed Residual Attention Network is a convolutional neural network using attention mechanism which can incorporate with state-of-art feed forward network architecture in an end-to-end training fashion and can be easily scaled up to hundreds of layers.
- DeeperForensics-1.0: A Large-Scale Dataset for Real-World Face Forgery Detection648
The on-going effort of constructing a large- scale benchmark for face forgery detection is presented, with 60, 000 videos constituted by a total of 17.6 million frames, 10 times larger than existing datasets of the same kind.
- Look at Boundary: A Boundary-Aware Face Alignment Algorithm523
A novel boundary-aware face alignment algorithm by utilising boundary lines as the geometric structure of a human face to help facial landmark localisation, which outperforms state-of-the-art methods by a large margin.
- MEAD: A Large-Scale Audio-Visual Dataset for Emotional Talking-Face Generation522
An emotional talking-face generation baseline that enables the manipulation of both emotion and its intensity and could be used for a number of research projects including conditional generation, cross-modal understanding and expression recognition.
- Bi-directional Cross-Modality Feature Propagation with Separation-and-Aggregation Gate for RGB-D Semantic Segmentation478
A unified and efficient Cross-modality Guided Encoder to not only effectively recalibrate RGB feature responses, but also to distill accurate depth information via multiple stages and aggregate the two recalibrated representations alternatively.
- TAM: Temporal Adaptive Module for Video Recognition388
A new temporal adaptive module (TAM) to generate video-specific temporal kernels based on its own feature map that outperforms other temporal modeling methods consistently, and achieves the state-of-the-art performance under the similar complexity.
- Whole-Body Human Pose Estimation in the Wild376
COCO-WholeBody is introduced which extends COCO dataset with whole-body annotations and is the first benchmark that has manual annotations on the entire human body, including 133 dense landmarks with 68 on the face, 42 on hands and 23 on the body and feet.
- PPDM: Parallel Point Detection and Matching for Real-Time Human-Object Interaction Detection337
A single-stage Human-Object Interaction (HOI) detection method that has outperformed all existing methods on HICO-DET dataset at 37 fps on a single Titan XP GPU is proposed, and is the first real-time HOI detection method.
- SimMatch: Semi-supervised Learning with Similarity Matching271
A new semi-supervised learning framework, SimMatch, is introduced, which simulta-neously considers semantic similarity and instance similarity and the unfolding and aggregation operation which allows these two similarities be isomorphically transformed with each other is proposed.
- A Real-Time Cross-Modality Correlation Filtering Method for Referring Expression Comprehension253
A novel Realtime Cross-modality Correlation Filtering method (RCCF) that reformulates the referring expression comprehension as a correlation ï¬ltering process and achieves leading performance in RefClef, RefCOCO, Ref COCO+ and RefC OCOg benchmarks.
- ReenactGAN: Learning to Reenact Faces via Boundary Transfer229
The proposed method, known as ReenactGAN, is capable of transferring facial movements and expressions from an arbitrary person's monocular video input to a target person’s video, and can perform photo-realistic face reenactment.
- A Coarse-to-Fine Adaptive Network for Appearance-Based Gaze Estimation219
A coarse-to-fine strategy which estimates a basic gaze direction from face image and refines it with corresponding residual predicted from eye images, and a framework which introduces a bi-gram model to bridge gaze residual andbasic gaze direction, and an attention component to adaptively acquire suitable fine-grained feature are proposed.
- The Devil of Face Recognition Is in the Noise217
This work contributes cleaned subsets of popular face databases, i.e., MegaFace and MS-Celeb-1M datasets, and builds a new large-scale noise-controlled IMDb-Face dataset, and investigates ways to improve data cleanliness, including a comprehensive user study on the influence of data labeling strategies to annotation accuracy.
- Deep Comprehensive Correlation Mining for Image Clustering204
A novel clustering framework, named deep comprehensive correlation mining~(DCCM), for exploring and taking full advantage of various kinds of correlations behind the unlabeled data from three aspects: Instead of only using pair-wise information, pseudo-label supervision is proposed to investigate category information and learn discriminative features.
- CentripetalNet: Pursuing High-Quality Keypoint Pairs for Object Detection193
This paper proposes CentripetalNet which uses centripetals shift to pair corner keypoints from the same instance and predicts the position and the centripeteal shift of the corner points and matches corners whose shifted results are aligned.
- Not All Tokens Are Equal: Human-centric Visual Analysis via Token Clustering Transformer192
A novel Vision Transformer, called Token Clustering Transformer (TCFormer), which merges tokens by progressive clustering, where the tokens can be merged from different locations with flexible shapes and sizes, which is beneficial to capturing detailed information.
- A Keypoint-based Global Association Network for Lane Detection191
A Global Association Network (GANet) is proposed to formulate the lane detection problem from a new perspective, where each keypoint is directly regressed to the starting point of the lane line instead of point-by-point extension, which outperforms previous methods.
- Reformulating HOI Detection as Adaptive Set Prediction190
This paper reformulate HOI detection as an adaptive set prediction problem, with this novel formulation, it is proposed an Adaptive Set-based one-stage framework (AS-Net) with parallel instance and interaction branches and an effective instance-aware attention module.
- Taming the Power of Diffusion Models for High-Quality Virtual Try-On with Appearance Flow187
The approach, namely Diffusion-based Conditional Inpainting for Virtual Try-ON(DCI-VTON), effectively utilizes the power of the diffusion model, and the incorporation of the warping module helps to produce high-quality and realistic virtual try-on results.
- Everybody’s Talkin’: Let Me Talk as You Want172
A method to edit a target portrait footage by taking a sequence of audio as input to synthesize a photo-realistic video, which is end-to-end learnable and robust to voice variations in the source audio.
- Weakly Supervised Contrastive Learning168
A weakly supervised contrastive learning framework (WCL) based on two projection heads, one of which will perform the regular instance discrimination task, and the other head will use a graph-based method to explore similar samples and generate a weak label to pull the similar images closer.
- GreedyNAS: Towards Fast One-Shot NAS With Greedy Supernet159
This paper proposes a multi-path sampling strategy with rejection, and greedily filter the weak paths to ease the burden of supernet by encouraging it to focus more on evaluation of those potentially-good ones, which are identified using a surrogate portion of validation data.
- Differentiable Hierarchical Graph Grouping for Multi-person Pose Estimation147
This paper proposes a novel differentiable Hierarchical Graph Grouping (HGG) method, which takes human keypoint candidates as graph nodes and clusters keypoints in a multi-layer graph neural network model and improves the performance of bottom-up pose estimation methods.
- Agree to Disagree: Adaptive Ensemble Knowledge Distillation in Gradient Space144
This paper examines the diversity of teacher models in the gradient space and regard the ensemble knowledge distillation as a multi-objective optimization problem so that the method can be seen as a dynamic weighting method for each teacher in the ensemble.
- 122
- Weakly-Supervised Discovery of Geometry-Aware Representation for 3D Human Pose Estimation115
A geometry-aware 3D representation for the human pose is proposed to address this limitation by using multiple views in a simple auto-encoder model at the training stage and only 2D keypoint information as supervision, and injecting the representation as a robust 3D prior.
- TransGaGa: Geometry-Aware Unsupervised Image-To-Image Translation114
A novel disentangle-and-translate framework to tackle the complex objects image-to-image translation task, which disentangles image space into a Cartesian product of the appearance and the geometry latent spaces and supports multimodal translation.
- Single image portrait relighting via explicit multiple reflectance channel modeling111
A novel framework that explicitly models multiple reflectance channels for single image portrait relighting, including the facial albedo, geometry as well as two lighting effects, i.e., specular and shadow is proposed.
- 3D Human Pose Machines with Self-supervised Learning97
A simple yet effective self-supervised correction mechanism to learn all intrinsic structures of human poses from abundant images and develops a 3D human pose machine, which jointly integrates the 2D spatial relationship, temporal smoothness of predictions and 3D geometric knowledge.
- Aggregation via Separation: Boosting Facial Landmark Detector With Semi-Supervised Style Translation95
A new perspective of facial landmark detection is investigated and it is demonstrated it leads to further notable improvement and surprisingly outperforms the fully-supervised model by a large margin.
- Multi-Person Articulated Tracking With Spatial and Temporal Embeddings93
A unified framework for multi-person pose estimation and tracking, which combines SpatialNet and TemporalNet, and model the grouping procedure into a differentiable Pose-Guided Grouping (PGG) module to make the whole part detection and grouping pipeline fully end-to-end trainable.
- HMOR: Hierarchical Multi-person Ordinal Relations for Monocular Multi-person 3D Pose Estimation87
This paper attempts to address the lack of a global perspective of the top-down approaches to 3D human pose estimation by introducing a novel form of supervision - Hierarchical Multi-person Ordinal Relations (HMOR).
- Make a Face: Towards Arbitrary High Fidelity Face Manipulation78
This work proposes Additive Focal Variational Auto-encoder (AF-VAE), a novel approach that can arbitrarily manipulate high-resolution face images using a simple yet effective model and only weak supervision of reconstruction and KL divergence losses.
- Graph-Based 3D Multi-Person Pose Estimation Using Multi-View Images74
This paper decomposes the task of estimating the 3D human poses of multiple persons from multiple calibrated camera views into two stages, i.e. person localization and pose estimation, and proposes three task-specific graph neural networks for effective message passing.
- When Human Pose Estimation Meets Robustness: Adversarial Algorithms and Benchmarks72
Extensive experiments show that AdvMix significantly increases the robustness of pose estimations across a wide range of corruptions, while maintaining accuracy on clean data in various challenging benchmark datasets.
- ViTAS: Vision Transformer Architecture Search69
This paper develops a new cyclic weight-sharing mechanism for token embeddings of the ViTs, which enables each channel could more evenly contribute to all candidate architectures, and proposes identity shifting to alleviate the many-to-one issue in superformer.
- TransMoMo: Invariance-Driven Unsupervised Video Motion Retargeting66
A lightweight video motion retargeting approach TransMoMo that is capable of transferring motion of a person in a source video realistically to another video of a target person by exploiting invariance properties of three orthogonal factors of variation including motion, structure, and view-angle.
- DRPose3D: Depth Ranking in 3D Human Pose Estimation66
This paper proposes a two-stage depth ranking based method that outperforms the state-of-the-art methods in the Human3.6M benchmark for all three testing protocols, indicating that depth ranking is an essential geometric feature which can be learned to improve the 3D pose estimation.
- ViPNAS: Efficient Video Pose Estimation via Neural Architecture Search64
A novel neural architecture search (NAS) method, termed ViP-NAS, to search networks in both spatial and temporal levels for fast online video pose estimation, which is the first to search for the temporal feature fusion and automatic computation allocation in videos.
- Video Semantic Segmentation via Sparse Temporal Transformer63
This work proposes a novel Sparse Temporal Transformer (STT) to bridge temporal relation among video frames adaptively, which is also equipped with query selection and key selection, and achieves the state-of-the-art segmentation accuracy and temporal consistency with comparable inference speed.
- 61
- Local Correlation Consistency for Knowledge Distillation60
To make the student focus on those informative local regions of the teacher's feature maps, a novel class-aware attention module is proposed to highlight the class-relevant regions and remove the confusing class-irrelevant regions, which makes the local correlation knowledge more accurate and valuable.
- Prioritized Architecture Sampling with Monto-Carlo Tree Search58
This paper introduces a sampling strategy based on Monte Carlo tree search (MCTS) with the search space modeled as a Monte Carlo trees (M CT), which captures the dependency among layers and significantly improves search efficiency and performance.
- Learning Where to Learn in Cross-View Self-Supervised Learning54
This paper reinterprets the projection head in SSL as a per-pixel projection and predicts a set of spatial alignment maps from the original features by this weight-sharing projection head, so that the projected embeddings could be exactly aligned and thus guide the feature learning better.
- Towards Improving the Consistency, Efficiency, and Flexibility of Differentiable Neural Architecture Search52
This paper introduces EnTranNAS that is composed of Engine-cells and Transit-cells, and develops an architecture derivation method to replace the traditional one that is based on a hand-crafted rule, which enables differentiable sparsification, and keeps the derived architecture equivalent to that of engines, which further improves the consistency between search and evaluation.
- 47
- A Cascaded Inception of Inception Network With Attention Modulated Feature Fusion for Human Pose Estimation46
Three novel techniques step by step to efficiently utilize different levels of features for human pose estimation are presented, designed to emphasize the low level features and a cascaded network is proposed to sequentially localize the joints to enforce message passing.
- 45
- 44
- BCNet: Searching for Network Width with Bilaterally Coupled Network42
This paper introduces a new supernet called Bilaterally Coupled Network (BCNet), which is fairly trained and responsible for the same amount of network widths, thus each network width can be evaluated more accurately and turns out to further boost the performance of NAS models by refining their networkwidths.
- 41
- 41
- FAB: A Robust Facial Landmark Detection Framework for Motion-Blurred Videos39
A framework named FAB is proposed that takes advantage of structure consistency in the temporal dimension for facial landmark detection in motion-blurred videos, and a structure predictor is proposed to predict the missing face structural information temporally, which serves as a geometry prior.
- 36
- 35
- Reborn Filters: Pruning Convolutional Neural Networks with Limited Data33
This paper proposes to use all original filters to directly develop new compact filters, named reborn filters, so that all useful structure priors in the original filters can be well preserved into the pruned networks, alleviating the performance drop accordingly.
- 33
- 32
- Amodal Instance Segmentation via Prior-Guided Expansion29
A prior-guided expansion framework is proposed, which builds on a two-stage segmentation model and performs box-level (resp., pixel-level) expansion for amodal box (resp, mask) prediction, by retrieving regression transformations from a memory bank of expansion prior.
- 29
- 29
- TCFormer: Visual Recognition via Token Clustering Transformer28
The Token Clustering Transformer (TCFormer), which generates dynamic vision tokens based on semantic meaning that represent image regions with similar semantic meanings using the same vision token, even if those regions are not adjacent.
- Progressive Attention on Multi-Level Dense Difference Maps for Generic Event Boundary Detection28
An effective and end-to-end learnable framework (DDM-Net) to tackle the diversity and complicated semantics of event boundaries, and demonstrates the effectiveness of richer motion representation and more sophisticated aggregation, in handling the diversity of GEBD.
- 27
- GreedyNASv2: Greedier Search with a Greedy Path Filter25
This paper leverages an explicit path filter to capture the characteristics of paths and directly filter those weak ones, so that the search can be thus implemented on the shrunk space more greedily and efficiently.
- 25
- An AI-Based Curling Game System for Winter Olympics24
CurlingHunter is the first, to the best of the knowledge, real-time system that can assist athletes to compete during the games in the history of sports and has been successfully applied in Winter Olympics and Winter Paralympics.
- 24
- 23
- DMVOS21
This work proposes Discriminative Matching for real-time Video Object Segmentation (DMVOS), a real- time VOS framework with high-accuracy to fill this gap in segmentation accuracy.
- 20
- 19
- TRB: A Novel Triplet Representation for Understanding 2D Human Body19
This paper proposes the Triplet Representation for Body --- a compact 2D human body representation, with skeleton keypoints capturing human pose information and contour keypoints containing human shape information, and proposes a two-branch network (TRB-net) with three novel techniques, namely X-structure, Directional Convolution and Pairwise mapping.
- 18
- Diffusing Winding Gradients (DWG): A Parallel and Scalable Method for 3D Reconstruction from Unoriented Point Clouds17
Diffusing Winding Gradients for reconstructing watertight surfaces from unoriented point clouds exploits the alignment between the gradients of the screened generalized winding number field—a robust variant of the standard GWN field—and globally consistent normals to orient points.
- 15
- 13
- Turbo Learning Framework for Human-Object Interactions Recognition and Human Pose Estimation13
A turbo learning framework to perform HOI recognition and pose estimation simultaneously and achieves the state-of-the-art performance on two public benchmarks including Verbs in COCO (V-COCO) and HICO-DET datasets.
- DeepFakes Detection: the DeeperForensics Dataset and Challenge11
This chapter presents the on-going effort of constructing DeeperForensics-1.0, a large-scale forgery detection dataset, and describes the observations obtained from organizing DeeperForensics Challenge 2020, a real-world face forgery detection competition based on DeeperForensics-1.0.
- 11
- GKGNet: Group K-Nearest Neighbor Based Graph Convolutional Network for Multi-label Image Recognition10
- Monocular Human Pose and Shape Reconstruction using Part Differentiable Rendering10
A part‐level differentiable renderer that enables part‐based models to be supervised by part segmentation in neural networks or optimization loops is proposed and a general parametric model engaged in the rendering pipeline as an intermediate representation between skeletons and detailed shapes is introduced for better interpretability.
- Talking Faces: Audio-to-Video Face Generation9
This chapter discusses the definition and underlying challenges of the talking face generation, and presents an overview of recent progress, as well as introducing some widely used datasets and performance metrics.
- 8
- 7
- 6
- Weak Augmentation Guided Relational Self-Supervised Learning6
This paper introduces a novel SSL paradigm, which is term as relational self-supervised learning (ReSSL) framework that learns representations by modeling the relationship between different instances by employing sharpened distribution of pairwise similarities among different instances as relation metric.
- Unlock the Power: Competitive Distillation for Multi-Modal Large Language Models6
An innovative Competitive Multi-modal Distillation framework (CoMD) is proposed, which captures bidirectional feedback between teacher and student models and continually updates the multi- modal capabilities that the student model has learned.
- 6
- 5
- TopoGen: Topology‐Aware 3D Generation with Persistence Points4
TopoGen is introduced, a method that extracts both discrete and continuous topological descriptors–Betti numbers and persistence points–using persistent homology, providing robust characterizations of 3D shapes in terms of their topology.
- SFDM: Robust Decomposition of Geometry and Reflectance for Realistic Face Rendering from Sparse-view Images4
A novel two-stage technique for decomposing and reconstructing facial features from sparse-view images, which outperforms existing methods by effectively disentangling geometric and reflectance components, significantly enhancing the quality of synthesized novel views, and paving the way for applications in facial relighting and reflectance editing.
- Parameterization-Driven Neural Surface Reconstruction for Object-Oriented Editing in Neural Rendering4
A novel neural algorithm for parameterizing neural implicit surfaces to simple parametric domains like spheres and polycubes that computes bi-directional deformation between the object and the domain using a forward mapping from the object's zero level set and an inverse deformation for backward mapping.
- 4
- Formulating facial mesh tracking as a differentiable optimization problem: a backpropagation-based solution3
This work proposes a backpropagation-based solution that formulates facial mesh tracking as a differentiable optimization problem called the BPMT, which exhibits a high degree of flexibility and extensibility, positioning it as a promising platform for future research in the community.
- 3
- 3
- Localin Reshuffle Net: Toward Naturally and Efficiently Facial Image Blending3
A novel local-reshuffle-layer is designed to map local patches in the feature space, which can be learned in an end-to-end fashion with dedicated loss and is essential for the proposed Local Reshuffle Network (LRNet).
- 2
- 2
- 2
- EllipBody: A Light-weight and Part-based Representation for Human Pose and Shape Recovery2
This work finds that part segmentation is a very efficient 2D annotation in 3D human body recovery and proposes a part-level differentiable renderer which model occlusion between parts explicitly, which enhances the performance in both learning-based and optimization-based methods.
- Knowledge-based hierarchical modeling evaluation method for skeleton models1
A hierarchical modeling and evaluation method for multi-level skeleton models based on the transitive closure is proposed and the effectiveness of the method is validated through the skeleton modeling and evaluation of the fairing on a rocket.
- Object Part Parsing with Hierarchical Dual Transformer1
The Hierarchical Dual Transformer (HDTR) is proposed to explore the contribution of the typical structural priors of the object parts to infer the hierarchical information from the part embeddings to enrich the pixel representations.
- 1
- 1
- 1
- –
- –
- Prior Metadata-Driven RAW Reconstruction: Eliminating the Need for Per-Image Metadata–
This work proposes using Prior Meta as a reference to reconstruct the RAW data instead of relying on per-image metadata, and achieves comparable, and even better, performance than RAW reconstruction methods based on metadata.
- –
- –
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 2026-10-15. 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.