VoxelTrack: Multi-Person 3D Human Pose Estimation and Tracking in the Wild

TL;DR

VoxelTrack introduces a voxel-based multi-view approach for robust multi-person 3D pose estimation and tracking, effectively handling occlusions and outperforming existing methods on multiple datasets.

Contribution

It proposes a novel voxel-based method that directly estimates and tracks 3D human poses from multi-view images without relying on noisy 2D pose correspondence.

Findings

Outperforms state-of-the-art on Shelf, Campus, and CMU Panoptic datasets.

Robustly handles severe occlusions in multi-view settings.

Avoids hard decisions based on individual images.

Abstract

We present VoxelTrack for multi-person 3D pose estimation and tracking from a few cameras which are separated by wide baselines. It employs a multi-branch network to jointly estimate 3D poses and re-identification (Re-ID) features for all people in the environment. In contrast to previous efforts which require to establish cross-view correspondence based on noisy 2D pose estimates, it directly estimates and tracks 3D poses from a 3D voxel-based representation constructed from multi-view images. We first discretize the 3D space by regular voxels and compute a feature vector for each voxel by averaging the body joint heatmaps that are inversely projected from all views. We estimate 3D poses from the voxel representation by predicting whether each voxel contains a particular body joint. Similarly, a Re-ID feature is computed for each voxel which is used to track the estimated 3D poses over time. The main advantage of the approach is that it avoids making any hard decisions based on individual images. The approach can robustly estimate and track 3D poses even when people are severely occluded in some cameras. It outperforms the state-of-the-art methods by a large margin on three public datasets including Shelf, Campus and CMU Panoptic.