Parts-Based Articulated Object Localization in Clutter Using Belief Propagation
This addresses the challenge for robots in human environments to perceive articulated objects like tools in clutter, though it appears incremental as it builds on existing parts-based and MRF approaches.
The paper tackles the problem of articulated object pose estimation in cluttered environments by proposing a generative-discriminative parts-based method using a Markov Random Field and belief propagation, achieving effective recognition and localization of hand tools in both uncluttered and cluttered tabletop settings.
Robots working in human environments must be able to perceive and act on challenging objects with articulations, such as a pile of tools. Articulated objects increase the dimensionality of the pose estimation problem, and partial observations under clutter create additional challenges. To address this problem, we present a generative-discriminative parts-based recognition and localization method for articulated objects in clutter. We formulate the problem of articulated object pose estimation as a Markov Random Field (MRF). Hidden nodes in this MRF express the pose of the object parts, and edges express the articulation constraints between parts. Localization is performed within the MRF using an efficient belief propagation method. The method is informed by both part segmentation heatmaps over the observation, generated by a neural network, and the articulation constraints between object parts. Our generative-discriminative approach allows the proposed method to function in cluttered environments by inferring the pose of occluded parts using hypotheses from the visible parts. We demonstrate the efficacy of our methods in a tabletop environment for recognizing and localizing hand tools in uncluttered and cluttered configurations.