Exposure-Midpoint Temporal Alignment and Risk-Aware Adaptive Feature Tracking for Stereo Visual–Inertial Odometry
Sucheng Yang, Qingqing Liu, Zhihong Zhuang, Xiaofeng Shen · 2026
Read Exposure-Midpoint Temporal Alignment and Risk-Aware Adaptive Feature Tracking for Stereo Visual–Inertial Odometry by Sucheng Yang, Qingqing Liu, Zhihong Zhuang, Xiaofeng Shen (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13611355.txt.
Description
Sucheng Yang and three co-authors improve how drones and robots track their own motion in a 2026 paper. Visual-inertial odometry fuses camera images with motion sensors, but automatic exposure changes and fast motion introduce timing errors and lost features. Their system rebuilds each image's true sampling moment at the midpoint of its exposure.
A risk model watching exposure, gain, and motion adjusts how features are found and followed in each frame. On indoor and outdoor data sets with known ground truth, the method cut position error by about a sixth to a quarter against baselines. It also ran faster than the standard system it extends, keeping real-time performance.
Catalog record
| CREATOR | Sucheng Yang, Qingqing Liu, Zhihong Zhuang, Xiaofeng Shen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 75 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |