SURABHI: Self-Training Using Rectified Annotations-Based Hard Instances for Eidetic Cattle Recognition

Manu Ramesh, Amy R Reibman, Zhaoyang Wang, Hieu Nguyen, John Hyatt, Minh P Vo · 2024

Read SURABHI: Self-Training Using Rectified Annotations-Based Hard Instances for Eidetic Cattle Recognition by Manu Ramesh, Amy R Reibman, Zhaoyang Wang, Hieu Nguyen, John Hyatt, Minh P Vo (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-11644942.txt.

Description

Ramesh and five co-authors present a training scheme for cattle recognition, in work from twenty twenty-four. Their system identifies cows from above by key points on the body. Exact key point spots are vital for naming the right cow.

The scheme, called SURABHI, trains the model on machine labeled cases it first got wrong. A rectifier fixes those hard cases, and the model learns from them. The team did not change the model shape itself. Gains came from better training cases. Final tests on cows the system had never seen showed a clear rise in correct IDs. Better ID can help farms track health and yield per animal.

Catalog record

CREATORManu Ramesh, Amy R Reibman, Zhaoyang Wang, Hieu Nguyen, John Hyatt, Minh P Vo
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE91 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC11644942

Browse the full library · About the Tower of Babel