Learning Normal Flow Directly From Event Neighborhoods
Dehao Yuan, Levi Burner, Jiayi Wu, Minghui Liu, Jingxi Chen · 2024
Read Learning Normal Flow Directly From Event Neighborhoods by Dehao Yuan, Levi Burner, Jiayi Wu, Minghui Liu, Jingxi Chen (2024) 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-2412-11284.txt.
Description
Dehao Yuan and co-authors present in 2024 a learning method that reads motion from event cameras, sensors that report only the pixels that change. Judging full motion from such sparse data is hard. Recent work settles for normal flow, the part of motion across an edge, which can be measured more reliably where texture is thin. Older estimators of it were model based and error prone.
Their method learns directly from the events around each point, using a small point cloud encoder. It gives sharp motion estimates in space and time. It accepts strong data augmentation, such as spinning the scene, which builds robustness. It can also report its own uncertainty, and trains on views that transfer between cameras. Tests showed better and steadier results than leading methods across data sets. A solver on its output, fused with motion sensors, handled hard cases well.
Catalog record
| CREATOR | Dehao Yuan, Levi Burner, Jiayi Wu, Minghui Liu, Jingxi Chen |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-11284