Dense Pose

Dense pose estimation aims to map each pixel of an image of a person to a corresponding 3D location on their body surface, providing a detailed, dense representation of human pose. Current research focuses on improving accuracy and efficiency, exploring various approaches including region-based convolutional neural networks (like R-CNN variants), transformer-based architectures for temporal upsampling and pose uplifting, and novel loss functions to stabilize training. These advancements are significant for applications such as human body reconstruction, motion capture, and human-computer interaction, offering more robust and efficient methods for analyzing human movement and appearance from various data sources, including images, videos, and even WiFi signals.

Papers

May 23, 2024

PoseCrafter: One-Shot Personalized Video Synthesis Following Flexible Pose Control
Yong Zhong, Min Zhao, Zebin You, Xiaofeng Yu, Changwang Zhang, Chongxuan Li
Human Pose Pose Control Dense Pose

March 16, 2024

DPPE: Dense Pose Estimation in a Plenoxels Environment using Gradient Approximation
Christopher Kolios, Yeganeh Bahoo, Sajad Saeedi
Neural Radiance Field Pose Estimation Neural Rendering Gradient Approximation Dense Pose

October 31, 2023

Pose-to-Motion: Cross-Domain Motion Retargeting with Pose Prior
Qingqing Zhao, Peizhuo Li, Wang Yifan, Olga Sorkine-Hornung, Gordon Wetzstein
Motion Synthesis Motion Transfer Pose Transfer Future Motion Dense Pose

August 19, 2023

Root Pose Decomposition Towards Generic Non-rigid 3D Reconstruction with Monocular Videos
Yikai Wang, Yinpeng Dong, Fuchun Sun, Xiao Yang
3D Reconstruction Monocular Video Non Rigid Dense Pose

December 31, 2022

DensePose From WiFi
Jiaqi Geng, Dong Huang, Fernando De la Torre
Computer Vision Mid Range LiDAR Human Sensing Automatic WiFi Dense Pose Densepose Model

November 4, 2022

UV R-CNN: Stable and Efficient Dense Human Pose Estimation
Wenhe Jia, Yilin Zhou, Xuhan Zhu, Mengjie Hu, Chun Liu, Qing Song
Point Prediction Dense Pose Densepose Model

October 12, 2022

Uplift and Upsample: Efficient 3D Human Pose Estimation with Uplifting Transformers
Moritz Einfalt, Katja Ludwig, Rainer Lienhart
3D Human Pose Estimation Monocular 3D Upsampling Operation Upsampling Module Pose Induced Video Transformer Dense Pose Transformer Era

July 6, 2022

DPODv2: Dense Correspondence-Based 6 DoF Pose Estimation
Ivan Shugurov, Sergey Zakharov, Slobodan Ilic
RGB Image Dense Correspondence Pose Optimization Dense Pose

April 4, 2022

Direct Dense Pose Estimation
Liqian Ma, Lingjie Liu, Christian Theobalt, Luc Van Gool
Human Pose Estimation Dense Correspondence Pose Transfer Dense Pose