3D Multi-Person Pose Estimation
18 papers with code • 5 benchmarks • 4 datasets
This task aims to solve root-relative 3D multi-person pose estimation. No human bounding box and root joint coordinate groundtruth are used in testing time.
( Image credit: RootNet )
LibrariesUse these libraries to find 3D Multi-Person Pose Estimation models and implementations
Although significant improvement has been achieved recently in 3D human pose estimation, most of the previous methods only treat a single-person case.
In contrast to the previous efforts which require to establish cross-view correspondence based on noisy and incomplete 2D pose estimations, we present an end-to-end solution which directly operates in the $3$D space, therefore avoids making incorrect decisions in the 2D space.
Our approach is self-improving by nature, since better network estimates can lead the optimization to better solutions, while more accurate optimization fits provide better supervision for the network.
Our method enables a realtime online motion capture system running at 30fps using 5 cameras on a 5-person scene.
To further verify the scalability of our method, we propose a new large-scale multi-human dataset with 12 to 28 camera views.
Our approach not only generalizes to in-the-wild images, but also yields a superior trade-off between speed and performance, compared to prior top-down approaches.
Through a body-center-guided sampling process, the body mesh parameters of all people in the image are easily extracted from the Mesh Parameter map.
In multi-person pose estimation actors can be heavily occluded, even become fully invisible behind another person.