Event
B-Adapt and BVSLAM improve SLAM robustness under rapid illumination changes
1 reports · 1 sources
Olivier Gamache and co-authors propose B-Adapt, an adaptive multi-exposure bracketing method, and BVSLAM, a multi-exposure visual SLAM method, addressing the problem that under rapid illumination changes, auto-exposure does not adjust in time and image saturation causes visual SLAM to lose correspondences. B-Adapt captures adaptive exposure brackets to cover the scene's dynamic range while maintaining photometric overlap; BVSLAM directly uses the multi-exposure sequence as constraints in factor-graph state estimation and does not rely on HDR synthesis. The authors report that in tests simulating a vehicle speed of 32 m/s with repeated tunnel-exit transitions, combining B-Adapt with BVSLAM inside the tunnel reduced state estimation error by 78% compared with auto-exposure images. The code has been made publicly available.
Report timeline
- B-Adapt and BVSLAM: Adaptive Multi-Exposure Bracketing for More Robust Visual SLAM
The authors propose B-Adapt and BVSLAM to address image saturation and lost visual correspondences in visual SLAM when automatic exposure fails to adjust in time under rapid illumination changes. B-Adapt captures adaptive exposure brackets to cover the scene's dynamic range while maintaining photometric overlap, and BVSLAM uses the multi-exposure sequence directly as constraints in factor-graph state estimation without relying on HDR merging.
arXiv cs.RO ·