Bodies at Rest: 3D Human Pose and Shape Estimation from a Pressure Image using Synthetic Data

People spend a substantial part of their lives at rest in bed. 3D human pose\nand shape estimation for this activity would have numerous beneficial\napplications, yet line-of-sight perception is complicated by occlusion from\nbedding. Pressure sensing mats are a promising alternative, but training data\nis challenging to collect at scale. We describe a physics-based method that\nsimulates human bodies at rest in a bed with a pressure sensing mat, and\npresent PressurePose, a synthetic dataset with 206K pressure images with 3D\nhuman poses and shapes. We also present PressureNet, a deep learning model that\nestimates human pose and shape given a pressure image and gender. PressureNet\nincorporates a pressure map reconstruction (PMR) network that models pressure\nimage generation to promote consistency between estimated 3D body models and\npressure image input. In our evaluations, PressureNet performed well with real\ndata from participants in diverse poses, even though it had only been trained\nwith synthetic data. When we ablated the PMR network, performance dropped\nsubstantially.\n