3D Human Pose Estimation
3D Human Pose Estimation is a task in the field of computer vision that aims to estimate the three-dimensional positions and orientations of human joints and bones from two-dimensional images or videos. The goal of this technology is to achieve real-time 3D human pose reconstruction, which has a wide range of applications, including virtual reality, human-computer interaction, and motion analysis.
Human3.6M
TCPFormer
3DPW
WHAM (ViT)
MPI-INF-3DHP
UGCN
HumanEva-I
GLA-GCN (T=27, GT)
H3WB
RTMW3D-x
Total Capture
AdaFuse
EMDB
TRAM
AGORA
NIKI (Twist-and-Swing)
Panoptic
ROMP (ResNet-50)
Surreal
VirtualMarker
3D Poses in the Wild Challenge
MeTRo
AIST++
RobustCap
RICH
SLOPER4D
LiDAR-HMR
UBody
Multi-HMR
SkiPose
MetaPose (S1+S2)
DHP19
Point Transformer
Geometric Pose Affordance
SPEC-MTP
W-HMR
3DOH50K
CHALL H80K
ResNet
Geometric Pose Affordance
Baseline model
HSPACE
ITOP front-view
SPiKE
JTA
Dual network
Waymo Open Dataset
VoxelKP