Enhancing the Ground Truth Disparity by MAP Estimation for Developing a Neural-Net Based Stereoscopic Camera
Hanbit Gil, Sehyun Ryu, Sungmin Woo, Junxing Zheng, Peng Cao · 2024
Read Enhancing the Ground Truth Disparity by MAP Estimation for Developing a Neural-Net Based Stereoscopic Camera by Hanbit Gil, Sehyun Ryu, Sungmin Woo, Junxing Zheng, Peng Cao (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/pmc-11644956.txt.
Description
Hanbit Gil and co-authors improve depth maps for stereo cameras in a 2024 paper. Stereo depth maps guide new camera tools that learn from sample data. A classic method gives sharp depth where it works, but leaves holes at edges and plain areas.
The team fills those holes with a method that weighs nearby depth values and the odds of each guess. The fix keeps the sharp parts of the classic map and cleans up its look. Cameras trained on the fixed maps did better on both test scenes and real scenes. The method suits self-driving and other uses that need firm depth from two views.
Catalog record
| CREATOR | Hanbit Gil, Sehyun Ryu, Sungmin Woo, Junxing Zheng, Peng Cao |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 42 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |