Search Results for author:
Found 91 papers, 24 papers with code
Hand-Transformer: Non-Autoregressive Structured Modeling for 3D Hand Pose Estimation
To address this issue, we connect this structured output learning problem with the structured modeling framework in sequence transduction field.
Learning Progressive Joint Propagation for Human Motion Prediction
Last, in order to incorporate a general motion space for high-quality prediction, we build a memory-based dictionary, which aims to preserve the global motion patterns in training data to guide the predictions.
Clustering Driven Deep Autoencoder for Video Anomaly Detection
Because of the ambiguous definition of anomaly and the complexity of real data, anomaly detection in videos is one of the most challenging problems in intelligent video surveillance.
Progressive Multi-view Human Mesh Recovery with Self-Supervision
To date, little attention has been given to multi-view 3D human mesh estimation, despite real-life applicability (e. g., motion capture, sport analysis) and robustness to single-view ambiguities.
GRiT: A Generative Region-to-text Transformer for Object Understanding
Specifically, GRiT consists of a visual encoder to extract image features, a foreground object extractor to localize objects, and a text decoder to generate open-set object descriptions.
Ranked #1 on
Dense Captioning
on Visual Genome
NeRFPlayer: A Streamable Dynamic Scene Representation with Decomposed Neural Radiance Fields
Visually exploring in a real-world 4D spatiotemporal space freely in VR has been a long-term quest.
Federated Learning with Privacy-Preserving Ensemble Attention Distillation
We propose a privacy-preserving FL framework leveraging unlabeled public data for one-way offline knowledge distillation in this work.
PREF: Predictability Regularized Neural Motion Fields
We propose to regularize the estimated motion to be predictable.
PointACL:Adversarial Contrastive Learning for Robust Point Clouds Representation under Adversarial Attack
Adversarial contrastive learning (ACL) is considered an effective way to improve the robustness of pre-trained models.
Neural Correspondence Field for Object Pose Estimation
Unlike classical correspondence-based methods which predict 3D object coordinates at pixels of the input image, the proposed method predicts 3D object coordinates at 3D query points sampled in the camera frustum.
AiATrack: Attention in Attention for Transformer Visual Tracking
However, the independent correlation computation in the attention mechanism could result in noisy and ambiguous attention weights, which inhibits further performance improvement.
Ranked #1 on
Visual Object Tracking
on NeedForSpeed
Semantics-Depth-Symbiosis: Deeply Coupled Semi-Supervised Learning of Semantics and Depth
Multi-task learning (MTL) paradigm focuses on jointly learning two or more tasks, aiming for significant improvement w. r. t model's generalizability, performance, and training/inference memory footprint.
Optical Flow for Video Super-Resolution: A Survey
Video super-resolution is currently one of the most active research topics in computer vision as it plays an important role in many visual applications.
Deformable VisTR: Spatio temporal deformable attention for video instance segmentation
To improve the training efficiency, we propose Deformable VisTR, leveraging spatio-temporal deformable attention module that only attends to a small fixed set of key spatio-temporal sampling points around a reference point.
Efficient Video Instance Segmentation via Tracklet Query and Proposal
In addition, VisTR is not fully end-to-end learnable in multiple video clips as it requires a hand-crafted data association to link instance tracklets between successive clips.
MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human Pose Estimation in Video
Recent transformer-based solutions have been introduced to estimate 3D human pose from 2D keypoint sequence by considering body joints among all frames globally to learn spatio-temporal correlation.
Ranked #3 on
Monocular 3D Human Pose Estimation
on Human3.6M
Motion-driven Visual Tempo Learning for Video-based Action Recognition
Action visual tempo characterizes the dynamics and the temporal scale of an action, which is helpful to distinguish human actions that share high similarities in visual dynamics and appearance.
Ranked #7 on
Action Recognition
on Something-Something V1
Joint-bone Fusion Graph Convolutional Network for Semi-supervised Skeleton Action Recognition
In recent years, graph convolutional networks (GCNs) play an increasingly critical role in skeleton-based human action recognition.
Consistent 3D Hand Reconstruction in Video via self-supervised Learning
We present a method for reconstructing accurate and consistent 3D hands from a monocular video.
Learning Transferable Human-Object Interaction Detector With Natural Language Supervision
More specifically, we propose a new HOI visual encoder to detect the interacting humans and objects, and map them to a joint feature space to perform interaction recognition.
Pseudo Supervised Monocular Depth Estimation with Teacher-Student Network
Despite recent improvement of supervised monocular depth estimation, the lack of high quality pixel-wise ground truth annotations has become a major hurdle for further progress.
OVIS: Open-Vocabulary Visual Instance Search via Visual-Semantic Aligned Representation Learning
We introduce the task of open-vocabulary visual instance search (OVIS).
Two-Stream Consensus Network: Submission to HACS Challenge 2021 Weakly-Supervised Learning Track
The base model training encourages the model to predict reliable predictions based on single modality (i. e., RGB or optical flow), based on the fusion of which a pseudo ground truth is generated and in turn used as supervision to train the base models.
Optical Flow Estimation
Weakly-supervised Temporal Action Localization
+1
NeuLF: Efficient Novel View Synthesis with Neural 4D Light Field
In this paper, we present an efficient and robust deep learning solution for novel view synthesis of complex scenes.
Weakly Supervised Temporal Action Localization Through Learning Explicit Subspaces for Action and Context
To address this challenge, we introduce a framework that learns two feature subspaces respectively for actions and their context.
Action Recognition
Weakly-supervised Temporal Action Localization
+1
ACSNet: Action-Context Separation Network for Weakly Supervised Temporal Action Localization
In this paper, we introduce an Action-Context Separation Network (ACSNet) that explicitly takes into account context for accurate action localization.
Ranked #6 on
Weakly Supervised Action Localization
on THUMOS’14
Video Polyp Segmentation
Weakly Supervised Action Localization
+2
Model-based 3D Hand Reconstruction via Self-Supervised Learning
For the first time, we demonstrate the feasibility of training an accurate 3D hand reconstruction network without relying on manual annotations.
Track to Detect and Segment: An Online Multi-Object Tracker
Most online multi-object trackers perform object detection stand-alone in a neural net without any input from tracking.
Ranked #1 on
Instance Segmentation
on nuScenes
Generation For Adaption: A GAN-Based Approach for 3D Domain Adaption with Point Cloud Data
Recent deep networks have achieved good performance on a variety of 3d points classification tasks.
Rethinking Soft Labels for Knowledge Distillation: A Bias-Variance Tradeoff Perspective
The outputs from the teacher network are used as soft labels for supervising the training of a new network.
Ranked #10 on
Knowledge Distillation
on ImageNet
SPAGAN: Shortest Path Graph Attention Network
SPAGAN therefore allows for a more informative and intact exploration of the graph structure and further {a} more effective aggregation of information from distant neighbors into the center node, as compared to node-based GCN methods.
A Unified 3D Human Motion Synthesis Model via Conditional Variational Auto-Encoder
Notably, by considering this problem as a conditional generation process, we estimate a parametric distribution of the missing regions based on the input conditions, from which to sample and synthesize the full motion series.
High Quality Disparity Remapping With Two-Stage Warping
In the second stage, we derive another warping model to refine warping results in less important regions by eliminating serious distortions in shape, disparity and 3D structure.
Rethinking Soft Labels for Knowledge Distillation: A Bias–Variance Tradeoff Perspective
In this paper, we investigate the bias-variance tradeoff brought by distillation with soft labels.
Discovering Human Interactions With Large-Vocabulary Objects via Query and Multi-Scale Detection
In this work, we study the problem of human-object interaction (HOI) detection with large vocabulary object categories.
Stacked Homography Transformations for Multi-View Pedestrian Detection
This task is confronted with two challenges: how to establish the 3D correspondences from views to the BEV map and how to assemble occupancy information across views.
Interventional Domain Adaptation
To address this issue, we intervene in the learning of feature discriminability using unlabeled target data to guide it to get rid of the domain-specific part and be safely transferable.
Two-Stream Consensus Network for Weakly-Supervised Temporal Action Localization
Weakly-supervised Temporal Action Localization (W-TAL) aims to classify and localize all action instances in an untrimmed video under only video-level supervision.
Ranked #10 on
Weakly Supervised Action Localization
on THUMOS14
Weakly Supervised Action Localization
Weakly-supervised Temporal Action Localization
+1
Attention-Aware Noisy Label Learning for Image Classification
In this paper, the attention-aware noisy label learning approach ($A^2NL$) is proposed to improve the discriminative capability of the network trained on datasets with potential label noise.
ConsNet: Learning Consistency Graph for Zero-Shot Human-Object Interaction Detection
We consider the problem of Human-Object Interaction (HOI) Detection, which aims to locate and recognize HOI instances in the form of <human, action, object> in images.
Human-Object Interaction Detection
Zero-Shot Human-Object Interaction Detection
Forest R-CNN: Large-Vocabulary Long-Tailed Object Detection and Instance Segmentation
In the classification tree, as the number of parent class nodes are significantly less, their logits are less noisy and can be utilized to suppress the wrong/noisy logits existed in the fine-grained class nodes.
Ranked #5 on
Few-Shot Object Detection
on LVIS v1.0 val
Revisiting Modified Greedy Algorithm for Monotone Submodular Maximization with a Knapsack Constraint
Second, we enhance the modified greedy algorithm to derive a data-dependent upper bound on the optimum.
Campus3D: A Photogrammetry Point Cloud Benchmark for Hierarchical Understanding of Outdoor Scene
Based on it, we formulate a hierarchical learning problem for 3D point cloud segmentation and propose a measurement evaluating consistency across various hierarchies.
Deep Reinforcement Learning with Label Embedding Reward for Supervised Image Hashing
Deep hashing has shown promising results in image retrieval and recognition.
Temporal Distinct Representation Learning for Action Recognition
Motivated by the previous success of Two-Dimensional Convolutional Neural Network (2D CNN) on image recognition, researchers endeavor to leverage it to characterize videos.
Structure-Aware Human-Action Generation
Generating long-range skeleton-based human actions has been a challenging problem since small deviations of one frame can cause a malformed action sequence.
Ranked #2 on
Human action generation
on NTU RGB+D 2D
Joint Hand-object 3D Reconstruction from a Single Image with Cross-branch Feature Fusion
In this work, we propose to consider hand and object jointly in feature space and explore the reciprocity of the two branches.
Towards Understanding the Adversarial Vulnerability of Skeleton-based Action Recognition
We first formulate generation of adversarial skeleton actions as a constrained optimization problem by representing or approximating the physiological and physical constraints with mathematical formulations.
3DV: 3D Dynamic Voxel for Action Recognition in Depth Video
Each available 3DV voxel intrinsically involves 3D spatial and motion feature jointly.
Image Co-skeletonization via Co-segmentation
Object skeletonization in a single natural image is a challenging problem because there is hardly any prior knowledge about the object.
Measuring Generalisation to Unseen Viewpoints, Articulations, Shapes and Objects for 3D Hand Pose Estimation under Hand-Object Interaction
To address these issues, we designed a public challenge (HANDS'19) to evaluate the abilities of current 3D hand pose estimators (HPEs) to interpolate and extrapolate the poses of a training set.
Temporal Pulses Driven Spiking Neural Network for Fast Object Recognition in Autonomous Driving
Accurate real-time object recognition from sensory data has long been a crucial and challenging task for autonomous driving.
Learning Diverse Stochastic Human-Action Generators by Learning Smooth Latent Transitions
In this paper, we focus on skeleton-based action generation and propose to model smooth and diverse transitions on a latent space of action sequences with much lower dimensionality.
Ranked #4 on
Human action generation
on NTU RGB+D 2D
A2J: Anchor-to-Joint Regression Network for 3D Articulated Pose Estimation from a Single Depth Image
For 3D hand and body pose estimation task in depth image, a novel anchor-based approach termed Anchor-to-Joint regression network (A2J) with the end-to-end learning ability is proposed.
Ranked #1 on
Hand Pose Estimation
on K2HPD
Context-Integrated and Feature-Refined Network for Lightweight Object Parsing
The core components of CIFReNet are the Long-skip Refinement Module (LRM) and the Multi-scale Context Integration Module (MCIM).
Bayesian Uncertainty Matching for Unsupervised Domain Adaptation
By imposing distribution matching on both features and labels (via uncertainty), label distribution mismatching in source and target data is effectively alleviated, encouraging the classifier to produce consistent predictions across domains.
Kervolutional Neural Networks
Convolutional neural networks (CNNs) have enabled the state-of-the-art performance in many computer vision tasks.
3D Hand Shape and Pose Estimation from a Single RGB Image
This work addresses a novel and challenging problem of estimating the full 3D hand shape and pose from a single RGB image.
Progress Regression RNN for Online Spatial-Temporal Action Localization in Unconstrained Videos
Previous spatial-temporal action localization methods commonly follow the pipeline of object detection to estimate bounding boxes and labels of actions.
Towards Real-time Eyeblink Detection in The Wild:Dataset,Theory and Practices
Effective and real-time eyeblink detection is of wide-range applications, such as deception detection, drive fatigue detection, face anti-spoofing, etc.
PointCloud Saliency Maps
Our motivation for constructing a saliency map is by point dropping, which is a non-differentiable operator.
Exploiting Local Feature Patterns for Unsupervised Domain Adaptation
In this paper, we present a method for learning domain-invariant local feature patterns and jointly aligning holistic and local feature statistics.
Bi-box Regression for Pedestrian Detection and Occlusion Estimation
The full body estimation branch is trained to regress full body regions for positive pedestrian proposals, while the visible part estimation branch is trained to regress visible part regions for both positive and negative pedestrian proposals.
Deformable Pose Traversal Convolution for 3D Action and Gesture Recognition
This deformable convolution can better utilize contextual joints for action and gesture recognition and is more robust to noisy joints.
Weakly-supervised 3D Hand Pose Estimation from Monocular RGB Images
Compared with depth-based 3D hand pose estimation, it is more challenging to infer 3D hand pose from monocular RGB images, due to substantial depth ambiguity and the difficulty of obtaining fully-annotated training data.
Point-to-Point Regression PointNet for 3D Hand Pose Estimation
Convolutional Neural Networks (CNNs)-based methods for 3D hand pose estimation with depth cameras usually take 2D depth images as input and directly regress holistic 3D hand pose.
Product Quantization Network for Fast Image Retrieval
Product quantization has been widely used in fast image retrieval due to its effectiveness of coding high-dimensional visual features.
Actor-Action Semantic Segmentation with Region Masks
One major challenge for this task is that when an actor performs an action, different body parts of the actor provide different types of cues for the action category and may receive inconsistent action labeling when they are labeled independently.
Recognizing Human Actions as the Evolution of Pose Estimation Maps
Specifically, the evolution of pose estimation maps can be decomposed as an evolution of heatmaps, e. g., probabilistic maps, and an evolution of estimated 2D human poses, which denote the changes of body shape and body pose, respectively.
Ranked #1 on
Multimodal Activity Recognition
on UTD-MHAD
Salience Guided Depth Calibration for Perceptually Optimized Compressive Light Field 3D Display
Multi-layer light field displays are a type of computational three-dimensional (3D) display which has recently gained increasing interest for its holographic-like effect and natural compatibility with 2D displays.
Conditional Generative Adversarial Network for Structured Domain Adaptation
In recent years, deep neural nets have triumphed over many computer vision problems, including semantic segmentation, which is a critical task in emerging autonomous driving and medical image diagnostics applications.
Hand PointNet: 3D Hand Pose Estimation Using Point Sets
Convolutional Neural Network (CNN) has shown promising results for 3D hand pose estimation in depth images.
Ranked #6 on
Hand Pose Estimation
on HANDS 2017
Multi-View Harmonized Bilinear Network for 3D Object Recognition
View-based methods have achieved considerable success in $3$D object recognition tasks.
Depth-Based 3D Hand Pose Estimation: From Current Achievements to Future Goals
Official Torch7 implementation of "V2V-PoseNet: Voxel-to-Voxel Prediction Network for Accurate 3D Hand and Human Pose Estimation from a Single Depth Map", CVPR 2018
Ranked #4 on
Hand Pose Estimation
on HANDS 2017
Non-iterative RGB-D-inertial Odometry
This paper presents a non-iterative solution to RGB-D-inertial odometry system.
Robotics
Multi-Label Learning of Part Detectors for Heavily Occluded Pedestrian Detection
Detecting pedestrians that are partially occluded remains a challenging problem due to variations and uncertainties of partial occlusion patterns.
Compressive Quantization for Fast Object Instance Search in Videos
Most of current visual search systems focus on image-to-image (point-to-point) search such as image and object retrieval.
Common Action Discovery and Localization in Unconstrained Videos
Similar to common object discovery in images or videos, it is of great interests to discover and locate common actions in videos, which can benefit many video analytics applications such as video summarization, search, and understanding.
Kernel Cross-Correlator
Cross-correlator plays a significant role in many visual perception tasks, such as object detection and tracking.
Spatio-Temporal Naive-Bayes Nearest-Neighbor (ST-NBNN) for Skeleton-Based Action Recognition
Moreover, by identifying key skeleton joints and temporal stages for each action class, our ST-NBNN can capture the essential spatio-temporal patterns that play key roles of recognizing actions, which is not always achievable by using end-to-end models.
HOPE: Hierarchical Object Prototype Encoding for Efficient Object Instance Search in Videos
This paper tackles the problem of efficient and effective object instance search in videos.
Object Co-Skeletonization With Co-Segmentation
Recent advances in the joint processing of images have certainly shown its advantages over the individual processing.
Fried Binary Embedding for High-Dimensional Visual Features
We argue that long binary codes (b O(d)) are critical to fully utilize the discriminative power of high-dimensional visual features, and can achieve better results in various tasks such as approximate nearest neighbour search.
3D Convolutional Neural Networks for Efficient and Robust Hand Pose Estimation From Single Depth Images
We propose a simple, yet effective approach for real-time hand pose estimation from single depth images using three-dimensional Convolutional Neural Networks (3D CNNs).
Robust 3D Hand Pose Estimation in Single Depth Images: from Single-View CNN to Multi-View CNNs
Articulated hand pose estimation plays an important role in human-computer interaction.
From Keyframes to Key Objects: Video Summarization by Representative Object Proposal Selection
This representative selection problem is formulated as a sparse dictionary selection problem, i. e., choosing a few representatives object proposals to reconstruct the whole proposal pool.
Adaptive Exponential Smoothing for Online Filtering of Pixel Prediction Maps
We propose an efficient online video filtering method, called adaptive exponential filtering (AES) to refine pixel prediction maps.
Fast Action Proposals for Human Action Detection and Search
Assuming each action is performed by a human with meaningful motion, both appearance and motion cues are utilized to measure the actionness of the video tubes.
Multi-feature Spectral Clustering with Minimax Optimization
To find a consensus clustering result that is agreeable to all feature modalities, our objective is to find a universal feature embedding, which not only fits each individual feature modality well, but also unifies different feature modalities by minimizing their pairwise disagreements.
Topical Video Object Discovery from Key Frames by Modeling Word Co-occurrence Prior
We show that such data driven co-occurrence information from bottom-up can conveniently be incorporated in LDA with a Gaussian Markov prior, which combines top down probabilistic topic modeling with bottom up priors in a unified model.
Max-Margin Structured Output Regression for Spatio-Temporal Action Localization
The mapping between a video and a spatio-temporal action trajectory is learned.
