RayMap3R: Inference-Time RayMap for Dynamic 3D Reconstruction
Feiran Wang, Zezhou Shang, Gaowen Liu, Yan Yan 1,† · 2026
Read RayMap3R: Inference-Time RayMap for Dynamic 3D Reconstruction by Feiran Wang, Zezhou Shang, Gaowen Liu, Yan Yan 1,† (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/arxiv-2603-20588.txt.
Description
Wang and co-authors present RayMap3R for streaming 3D reconstruction of dynamic scenes. Feed forward models can drift when objects move. Their method needs no new training at test time. RayMap outputs tend to favor static scenes, which gives a useful cue. A dual branch scheme contrasts RayMap and image outputs to spot motion. It then blocks those zones from memory updates. Reset alignment and state aware smoothing keep scale and paths steady.
Catalog record
| CREATOR | Feiran Wang, Zezhou Shang, Gaowen Liu, Yan Yan 1,† |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 54757 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.20588