OmniShape: Zero-Shot Multi-Hypothesis Shape and Pose Estimation in the Real World

-- We would like to estimate the pose and full shape of an object from a single observation, without assuming known 3D model or category. In this work, we propose OmniShape, the first method of its kind to enable probabilistic pose and shape estimation. OmniShape is based on the key insight that shape completion can be decoupled into two multi-modal distributions: one capturing how measurements project into a normalized object reference frame defined by the dataset and the other modelling a prior over object geometries represented as triplanar neural fields. By training separate conditional diffusion models for these two distributions, we enable sampling multiple hypotheses from the joint pose and shape distribution. OmniShape demonstrates compelling performance on challenging real world datasets. Detailed understanding of the 3D world is a core challenge in applications ranging from augmented reality to robotics. Despite recent progress in open-world image understanding [1], [2], estimating the complete and accurate 3D geometry of objects in a scene from a single view is an open problem.