Visual deep learning approaches for alphabetic sign language interpretation
Joyonta Kumar Nath, S A Mahadi Alam, Yasir Ullah, Miso Dhar, Joy Dewanjee, Muhammad Umair, Shahid Kamal, Zul Hilmi Abdullah · 2026
Read Visual deep learning approaches for alphabetic sign language interpretation by Joyonta Kumar Nath, S A Mahadi Alam, Yasir Ullah, Miso Dhar, Joy Dewanjee, Muhammad Umair, Shahid Kamal, Zul Hilmi Abdullah (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/PMC13522624.txt.
Description
Joyonta Kumar Nath and colleagues compare deep learning models for reading sign language alphabets from images in a 2026 study. Recognizing single-letter hand signs is a basic step toward accessible communication tools for people with hearing or speech impairment. Four pretrained image networks were tested under one protocol on a benchmark American Sign Language alphabet dataset.
The largest modern network reached the highest accuracy, just under ninety-nine point nine percent, with the other three close behind. Transfer learning, meaning reuse of networks pretrained on general images, proved effective at capturing fine hand shape differences. The comparison offers practical guidance for choosing models when building reliable, scalable sign language interpretation systems for real-world use.
Catalog record
| CREATOR | Joyonta Kumar Nath, S A Mahadi Alam, Yasir Ullah, Miso Dhar, Joy Dewanjee, Muhammad Umair, Shahid Kamal, Zul Hilmi Abdullah |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 53277 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |