Pose Estimation
2D Pose, tracking
3D Pose
- Camera Distortion-aware 3D Human Pose Estimation in Videowith Optimization-based Meta-Learning
- ElePose: Unsupervised 3D Human Pose Estimation by Predicting Camera Elevation and Learning Normalizing Flows on 2D Poses
- TesseTrack: End-to-End Learnable Multi-Person Articulated 3D Pose Tracking
SMPL
- Shape-aware Multi-Person Pose Estimation from Multi-View Images
- CVPR2022,Physics-aware Real-time Human Motion Tracking from Sparse Inertial Sensors: 从inertial传感器获得姿态
- CVPR2022, Capturing Humans in Motion: Temporal-Attentive 3D Human Pose and Shape Estimation from Monocular Video
- HSC4D: Human-centered 4D Scene Capture in Large-scale Indoor-outdoor Space Using Wearable IMUs and LiDAR
Performance Capture
- Human Performance Capture from Monocular Video in the Wild
- High-Fidelity Human Avatars from a Single RGB Camera
Hand Pose
- Local and Global Point Cloud Reconstruction for 3D Hand Pose Estimation
- Semi-Supervised 3D Hand Shape and PoseEstimation with Label Propagation
- Constraining Dense Hand Surface Tracking with Elasticity
- HybridCap: Inertia-aid Monocular Capture of Challenging Human Motions
- CVPR22: HandOccNet: Occlusion-Robust 3D Hand Mesh Estimation Network
-
dataset
- InterHand2.6M
- Capturing Hands in Action using Discriminative Salient Points and Physics Simulation
Human-Object
2D
- Improving Human-Object Interaction Detection via Phrase Learning and Label Composition: HOI detection
- Decoupling Object Detection from Human-Object Interaction Recognition
3D
Garment
- Robust 3D Garment Digitization from Monocular 2D Images for 3D VirtualTry-On Systems
- ICON: Implicit Clothed humans Obtained from Normals
Body Representation
- HVH: Learning a Hybrid Neural Volumetric Representation for Dynamic Hair Performance Capture
- Learning to Fit Morphable Models
- LatentHuman: Shape-and-Pose Disentangled Latent Representationfor Human Bodies
- Learning Body-Aware 3D Shape Generative Models
Implict head/body model
I M Avatar: Implicit Morphable Head Avatars from Videos
Performance Capture
- Siggraph18: MonoPerfCap: Human Performance Capture from Monocular Video; home
- CVPR20: DeepCap: Monocular Human Performance Capture Using Weak Supervision: 训练一个泛化的网络，给定一个人的重建的模型，输入一张分割好的图像，通过PoseNet和DefNet输出关节旋转角度与节点形变，获得mesh的形变；训练的时候使用关键点，轮廓进行监督.