Streaming mapping system improves long term accuracy and corrects errors
SURE-Map: Self-Correcting Streaming Geometric Foundation Model
Computer Vision and Pattern Recognition
Summary
Mapping systems that build 3D maps while moving face a problem: they see only a little bit at a time, so small mistakes add up and cause big errors over time. The authors propose SURE-Map, a method that not only predicts positions and depths but also checks itself for mistakes across multiple views. By combining fast updates with occasional deeper checks that fix accumulated errors, it keeps maps more accurate over long distances. Their system performs better than previous ones on several driving datasets.
What this means in practice
- •For autonomous vehicle engineers: Improve real-time 3D mapping accuracy over long drives by detecting and correcting cumulative mapping errors during vehicle navigation.
- •For robotics developers: Develop robots with enhanced self-correcting navigation systems that maintain geometric consistency in diverse environments despite dynamic obstacles.
Authors
Mingkai Liu, Hao Zhao, Xingxing Zuo
Abstract
Streaming geometric foundation models are emerging as a compelling alternative to SLAM systems. Yet this streaming nature introduces a fundamental issue: each prediction is made from limited context, which is vulnerable to dynamic objects and weak textures. Small local errors accumulate into severe geometric distortion and long-horizon scale drift. We argue that reliable streaming reconstruction requires geometric foundation models to be not only predictive, but also self-correcting. We introduce SURE-Map, a self-correcting framework built upon two complementary principles. First, we explicitly model cross-view geometric uncertainty. Unlike conventional depth or point confidence, which primarily reflects the reliability of individual-view prediction, our uncertainty directly measures whether the jointly predicted pose and depth induce geometrically consistent cross-view pixel correspondences. Second, because local correction alone cannot eliminate slowly accumulating scale errors, we introduce multi-timescale self-correction: fast consecutive-frame inference preserves streaming efficiency, while sparse keyframe-window inference provides longer-range geometric evidence to periodically recalibrate the scale of recent trajectories. SURE-Map establishes new state-of-the-art performance for online feed-forward reconstruction across long-horizon benchmarks, reducing ATE-RMSE from 24.00 to 17.24 m on KITTI, 5.11 to 4.74 m on Oxford Spires, and 31.37 to 28.58 m on VBR, with further improvements to 15.17, 4.63, and 22.12 m when incorporating loop-closure refinement. Project page: https://mingkai-liu.github.io/projects/sure-map/.