XNect: Real-time Multi-Person 3D Motion Capture with a Single RGB Camera

1 Jul 2019Dushyant MehtaOleksandr SotnychenkoFranziska MuellerWeipeng XuMohamed ElgharibPascal FuaHans-Peter SeidelHelge RhodinGerard Pons-MollChristian Theobalt

We present a real-time approach for multi-person 3D motion capture at over 30 fps using a single RGB camera. It operates successfully in generic scenes which may contain occlusions by objects and by other people... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
3D Human Pose Estimation Human3.6M SelecSLS Average MPJPE (mm) 63.6 # 36
3D Human Pose Estimation MPI-INF-3DHP XNect (SelecSLS) 3DPCK 82.8 # 9
AUC 45.3 # 10
MJPE 98.4 # 9
3D Multi-Person Pose Estimation MuPoTS-3D SelecSLS 3DPCK 75.8 # 1

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet