Weakly Supervised Multi-Modal 3D Human Body Pose Estimation for Autonomous Driving

TL;DR

This paper introduces a weakly supervised multi-modal approach for 3D human pose estimation in autonomous driving, leveraging camera and LiDAR data to improve accuracy without requiring extensive labeled datasets.

Contribution

The authors propose a novel weakly supervised method combining camera and LiDAR data for 3D human pose estimation in autonomous vehicles, reducing reliance on labeled data.

Findings

Outperforms state-of-the-art by up to 13% on Waymo dataset in weakly supervised setting

Achieves state-of-the-art results in fully supervised setting

Effective use of sensor fusion and pseudo labels for 3D pose estimation

Abstract

Accurate 3D human pose estimation (3D HPE) is crucial for enabling autonomous vehicles (AVs) to make informed decisions and respond proactively in critical road scenarios. Promising results of 3D HPE have been gained in several domains such as human-computer interaction, robotics, sports and medical analytics, often based on data collected in well-controlled laboratory environments. Nevertheless, the transfer of 3D HPE methods to AVs has received limited research attention, due to the challenges posed by obtaining accurate 3D pose annotations and the limited suitability of data from other domains. We present a simple yet efficient weakly supervised approach for 3D HPE in the AV context by employing a high-level sensor fusion between camera and LiDAR data. The weakly supervised setting enables training on the target datasets without any 2D/3D keypoint labels by using an off-the-shelf 2D joint extractor and pseudo labels generated from LiDAR to image projections. Our approach outperforms state-of-the-art results by up to $\sim$ 13% on the Waymo Open Dataset in the weakly supervised setting and achieves state-of-the-art results in the supervised setting.