Transformer-based iris verification with attention-guided segmentation and Siamese learning
S. Ramesh, V. Krishnaveni · 2026
Read Transformer-based iris verification with attention-guided segmentation and Siamese learning by S. Ramesh, V. Krishnaveni (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-13369665.txt.
Description
Ramesh and Krishnaveni present a new iris recognition system in this two thousand twenty-six paper. Checking identity by the eye's iris is common in secure settings. Their system blends several deep learning parts into one pipeline. A denoising autoencoder first cleans the image. An attention-guided network then pinpoints the iris region with care.
A vision transformer reads the iris texture as a whole, catching patterns across the full pattern. The system learns by comparing pairs of eyes, so that matches and non-matches pull apart. It was tested on a known iris data set with three subsets. The authors report a low error rate and strong accuracy scores. Such gains matter for border checks and secure access systems.
Catalog record
| CREATOR | S. Ramesh, V. Krishnaveni |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 97 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |