Author

Ezgi Ekin

h-index 0 View on Semantic Scholar

We have 1 of 1 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Aug 2026

Gloss-Free Representation Learning for Cross-Dataset Sign Spotting

The results show that loosely aligned broadcast data can provide effective weak supervision for learning sign representations that capture both lexical content and temporal structure, and raises top-5 temporal localization mean IoU from 0.235 to 0.465.

Oğuz Akif Tüfekcioğlu, Ezgi Ekin, Mustafa Kaan Çevik et al. · 0 citations