Integrating deep learning, biological hierarchies, and high-resolution imagery to create a new identification tool for cryptic coral reef fishes
Leonardo F Reginato, Simon J Brandl · 2026
Read Integrating deep learning, biological hierarchies, and high-resolution imagery to create a new identification tool for cryptic coral reef fishes by Leonardo F Reginato, Simon J Brandl (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/plos-pone-0349646.txt.
Description
Leonardo F Reginato and Simon J Brandl built a computer vision tool in 2026 for tiny reef fish that hide near the bottom. These fish are hugely varied and important, yet only a handful of experts can name them from photos, so most go uncounted.
Their system, CryptoVision, learns from high quality images and answers at three levels at once: family, genus, and species, respecting how those levels nest. A design that lifts weak image features helps it separate look-alike species. If the tool proves accurate in wider use, divers and camera surveys could track these hidden fish at a scale expert time alone could never reach.
Catalog record
| CREATOR | Leonardo F Reginato, Simon J Brandl |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 52728 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |