Synthetic Occlusion Augmentation with Volumetric Heatmaps for the 2018 ECCV PoseTrack Challenge on 3D Human Pose Estimation
This work addresses 3D human pose estimation for computer vision applications, with an incremental improvement through novel data augmentation.
The paper tackled 3D human pose estimation by using a fully-convolutional architecture with volumetric heatmaps and synthetic occlusion augmentation, achieving first place in the 2018 ECCV PoseTrack Challenge and surpassing state-of-the-art on the Human3.6M benchmark without additional pose datasets.
In this paper we present our winning entry at the 2018 ECCV PoseTrack Challenge on 3D human pose estimation. Using a fully-convolutional backbone architecture, we obtain volumetric heatmaps per body joint, which we convert to coordinates using soft-argmax. Absolute person center depth is estimated by a 1D heatmap prediction head. The coordinates are back-projected to 3D camera space, where we minimize the L1 loss. Key to our good results is the training data augmentation with randomly placed occluders from the Pascal VOC dataset. In addition to reaching first place in the Challenge, our method also surpasses the state-of-the-art on the full Human3.6M benchmark among methods that use no additional pose datasets in training. Code for applying synthetic occlusions is availabe at https://github.com/isarandi/synthetic-occlusion.