Heuristic Weakly Supervised 3D Human Pose Estimation in Novel Contexts without Any 3D Pose Ground Truth

Monocular 3D human pose estimation from a single RGB image has received a lot attentions in the past few year. Pose inference models with competitive performance however require supervision with 3D pose ground truth data or at least known pose priors in their target domain. Yet, these data requirements in many real-world applications with data collection constraints may not be attainable. In this paper, we present a heuristic weakly supervised human pose (HW-HuP) solution to estimate 3D human pose in contexts that no ground truth 3D pose data is accessible, even for fine-tuning. HW-HuP learns partial pose priors from public 3D human pose datasets and uses easy-to-access observations from the target domain to iteratively estimate 3D human pose and shape in an optimization and regression hybrid cycle. In our design, depth data as an auxiliary information is employed as weak supervision during training, yet it is not needed for the inference. HW-HuP shows comparable performance on public benchmarks to the state-of-the-art approaches which benefit from full 3D pose supervision. In this paper, we focus on two practical applications of 3D pose estimation for individuals while in bed as well as infants, where no reliable 3D pose data exists.