Compact Occlusion-Robust Facial Expression Recognition via Clean-Anchored Hard Occlusion Fine-Tuning
Xuefeng Zhao, Yixuan Dong, Zhaoman Zhong, Tingchen Jiang · 2026
Read Compact Occlusion-Robust Facial Expression Recognition via Clean-Anchored Hard Occlusion Fine-Tuning by Xuefeng Zhao, Yixuan Dong, Zhaoman Zhong, Tingchen Jiang (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-13568173.txt.
Description
Zhao and three co-authors teach a small net to read face moods through blocks in this twenty twenty-six study. Hands, shades, and masks can hide the face cues a mood read needs. Hard drill on fake blocks can fix that, yet it can also blunt reads on clean faces.
Inside, their drill hangs on clean faces as the fixed base for all tests. Fake blocks of set and random shapes are mixed in with face-zone weights. A frozen guide net from a prior stage keeps score spread steady through the drill. The end net scored high on both clean and blocked face sets in trials. Its size stays small, which suits live cam use on light gear.
Catalog record
| CREATOR | Xuefeng Zhao, Yixuan Dong, Zhaoman Zhong, Tingchen Jiang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 62 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |