Learning to Segment Dynamic Objects using SLAM Outliers
arXiv:2011.06259 · doi:10.1109/ICPR48806.2021.9412341
Abstract
We present a method to automatically learn to segment dynamic objects using SLAM outliers. It requires only one monocular sequence per dynamic object for training and consists in localizing dynamic objects using SLAM outliers, creating their masks, and using these masks to train a semantic segmentation network. We integrate the trained network in ORB-SLAM 2 and LDSO. At runtime we remove features on dynamic objects, making the SLAM unaffected by them. We also propose a new stereo dataset and new metrics to evaluate SLAM robustness. Our dataset includes consensus inversions, i.e., situations where the SLAM uses more features on dynamic objects that on the static background. Consensus inversions are challenging for SLAM as they may cause major SLAM failures. Our approach performs better than the State-of-the-Art on the TUM RGB-D dataset in monocular mode and on our dataset in both monocular and stereo modes.
arXiv commentPublished in the proceedings of ICPR 2020 (25th International Conference on Pattern Recognition): https://ieeexplore.ieee.org/abstract/document/9412341
References in corpus (5)
- ORB-SLAM2: an Open-Source SLAM System for Monocular, Stereo and RGB-D Cameras
- Past, Present, and Future of Simultaneous Localization And Mapping: Towards the Robust-Perception Age
- DS-SLAM: A Semantic Visual SLAM towards Dynamic Environments
- Ultimate SLAM? Combining Events, Images, and IMU for Robust Visual SLAM in HDR and High Speed Scenarios
- A Unified Framework for Mutual Improvement of SLAM and Semantic Segmentation