AI Paper Atlas

Research map / Video / Video segmentation and tracking

Neural Sign Language Translation

CVPR 2018 · Video › Video segmentation and tracking · approach: Sign language and physiological video analysis · 787 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add gloss-level intermediate supervision to direct video-to-text trainingSign language translation · PHOENIX14T testBLEU-4: 9.58 -> 18.13Helped
Train the translation network on predicted glosses rather than applying a ground-truth-gloss-trained modelS2G2T · PHOENIX14T testBLEU-4: 17.79 -> 18.13Helped
Use GRU units instead of LSTM units (ablation)G2T · PHOENIX14T testBLEU-4: 16.58 -> 18.75Helped
Add attention instead of using no attention (ablation)G2T · PHOENIX14T testBLEU-4: 16.34 -> 18.75Helped
Use Luong attention instead of Bahdanau attention (ablation)G2T · PHOENIX14T testBLEU-4: 17.40 -> 18.75Helped
Use batch size 1 instead of batch size 128 (ablation)G2T · PHOENIX14T testBLEU-4: 18.75 -> 19.26Helped
Add attention instead of using no attention (ablation)S2T · PHOENIX14T testBLEU-4: 8.35 -> 9.58Helped
Use Bahdanau attention instead of Luong attention (ablation)S2T · PHOENIX14T testBLEU-4: 9.00 -> 9.58Helped

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy