Geometry-Conditioned Diffusion for Occlusion-Robust In-Bed Pose Estimation

2026-04-26

AI Summary

This work addresses the challenge of insufficient robustness in bed-bound human pose estimation under blanket occlusion, primarily caused by the scarcity of annotated data. To this end, we propose Pose-LDM, a pose-conditioned latent diffusion model that formulates occlusion-aware data augmentation as a geometry generation task conditioned solely on sparse skeletal keypoints. Notably, our approach requires neither paired supervision nor pixel-level source image inputs, enabling—for the first time—the synthesis of realistic lying poses under arbitrary occlusions using only keypoint guidance. This significantly enhances pose diversity and system scalability. Experimental results demonstrate that, under severe occlusion, our method surpasses existing unpaired approaches in strict localization accuracy, achieves overall performance comparable to paired diffusion models, and closely approaches fully supervised training results.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Intelligent Robots: State Estimation

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems

Abstract

Robust in-bed human pose estimation under blanket occlusion remains challenging due to the scarcity of reliable labeled training data for heavily covered poses. Existing approaches rely on multi-modal sensing or image-to-image translation frameworks that remain conditioned on visible source imagery, limiting scalability and pose diversity. In this work, we reformulate occlusion-aware augmentation as a geometry-conditioned generative modeling task. We conduct a systematic comparison of deterministic masking, unpaired translation, paired diffusion-based translation, and a proposed pose-conditioned Latent Diffusion Model (Pose-LDM). Unlike image-guided methods, Pose-LDM synthesizes blanket-covered images directly from skeletal keypoints, eliminating dependence on paired supervision and pixel-level source-image conditioning while enabling generation from arbitrary pose inputs. All augmentation strategies are evaluated through their impact on downstream pose estimation under a fixed backbone. Pose- LDM achieves the highest strict localization accuracy under severe occlusion while maintaining overall detection performance comparable to paired diffusion models, approaching the performance of fully supervised training. These results demonstrate that geometry-conditioned diffusion provides an effective and supervision-efficient pathway toward occlusion-robust inbed pose estimation without modifying the sensing pipeline. The code is available at: github.com/navidTerraNova/ GeoDiffPose.

Problem

• in-bed pose estimation
• occlusion
• blanket occlusion
• pose estimation
• training data scarcity

Innovation

• geometry-conditioned diffusion
• occlusion-robust pose estimation
• Latent Diffusion Model
• blanket occlusion
• pose-conditioned generation