Multi-modal 3D Human Pose Estimation with 2D Weak Supervision in Autonomous Driving
This addresses the costly need for 3D data annotation in autonomous driving by enabling weak supervision from 2D labels, though it is incremental as it builds on existing multi-modal methods.
The paper tackles the problem of 3D human pose estimation in autonomous vehicles by proposing a multi-modal approach that uses 2D labels on RGB images as weak supervision, achieving a 22% relative improvement over a camera-only baseline and a 6% improvement over a LiDAR-only model on the Waymo Open Dataset.
3D human pose estimation (HPE) in autonomous vehicles (AV) differs from other use cases in many factors, including the 3D resolution and range of data, absence of dense depth maps, failure modes for LiDAR, relative location between the camera and LiDAR, and a high bar for estimation accuracy. Data collected for other use cases (such as virtual reality, gaming, and animation) may therefore not be usable for AV applications. This necessitates the collection and annotation of a large amount of 3D data for HPE in AV, which is time-consuming and expensive. In this paper, we propose one of the first approaches to alleviate this problem in the AV setting. Specifically, we propose a multi-modal approach which uses 2D labels on RGB images as weak supervision to perform 3D HPE. The proposed multi-modal architecture incorporates LiDAR and camera inputs with an auxiliary segmentation branch. On the Waymo Open Dataset, our approach achieves a 22% relative improvement over camera-only 2D HPE baseline, and 6% improvement over LiDAR-only model. Finally, careful ablation studies and parts based analysis illustrate the advantages of each of our contributions.