Persistent Homology Meets Object Unity: Object Recognition in Clutter
arXiv:2305.03815 · doi:10.1109/TRO.2023.3343994
Abstract
Recognition of occluded objects in unseen and unstructured indoor environments is a challenging problem for mobile robots. To address this challenge, we propose a new descriptor, TOPS, for point clouds generated from depth images and an accompanying recognition framework, THOR, inspired by human reasoning. The descriptor employs a novel slicing-based approach to compute topological features from filtrations of simplicial complexes using persistent homology, and facilitates reasoning-based recognition using object unity. Apart from a benchmark dataset, we report performance on a new dataset, the UW Indoor Scenes (UW-IS) Occluded dataset, curated using commodity hardware to reflect real-world scenarios with different environmental conditions and degrees of object occlusion. THOR outperforms state-of-the-art methods on both the datasets and achieves substantially higher recognition accuracy for all the scenarios of the UW-IS Occluded dataset. Therefore, THOR, is a promising step toward robust recognition in low-cost robots, meant for everyday use in indoor settings.
arXiv commentThis work has been accepted for publication in the IEEE Transactions on Robotics
References in corpus (7)
- PCT: Point cloud transformer
- Open3D: A Modern Library for 3D Data Processing
- Benchmarking in Manipulation Research: The YCB Object and Model Set and Benchmarking Protocols
- Human action recognition with a large-scale brain-inspired photonic computer
- Towards Domain Generalization in Object Detection
- Visual Object Recognition in Indoor Environments Using Topologically Persistent Features
- Domain Generalisation for Object Detection under Covariate and Concept Shift