AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

CVT-SLR: Contrastive Visual-Textual Transformation for Sign Language Recognition With Variational Alignment

Jiangbin Zheng, Yile Wang, Cheng Tan, Siyuan Li, Ge Wang, Jun Xia, Yidong Chen, Stan Z. Li

CVPR 2023 · Vision-language and multimodal › Cross-modal learning · approach: Contrastive vision-language alignment · 96 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (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 the pretrained textual module (ablation)CVT-SLR with pretrained visual module · PHOENIX-2014WER: 20.4 -> 19.8; 20.7 -> 20.1Improved
Add the pretrained textual module (ablation)CVT-SLR with pretrained visual module · PHOENIX-2014TWER: +1.7%/+1.6%Improved
Add the pretrained visual module (ablation)CVT-SLR with pretrained textual module · PHOENIX-2014WER: 30.8 -> 19.8; 31.4 -> 20.1Improved
Add contrastive alignment loss (ablation)CVT-SLR without pretrained VAE parameters · PHOENIX-2014WER: 21.1 -> 20.4; 21.4 -> 20.7Improved
Add pretrained VAE parameters (ablation)CVT-SLR without contrastive alignment · PHOENIX-2014WER: 21.1 -> 20.2; 21.4 -> 20.3Improved
Add pretrained VAE parameters (ablation)CVT-SLR without contrastive alignment · PHOENIX-2014TWER: 21.8 -> 20.1; 22.0 -> 20.4Improved
Add contrastive alignment loss (ablation)CVT-SLR with pretrained VAE parameters · PHOENIX-2014WER: 20.2 -> 19.8; 20.3 -> 20.1Improved
Add contrastive alignment loss (ablation)CVT-SLR with pretrained VAE parameters · PHOENIX-2014TWER: 20.1 -> 19.4; 20.4 -> 20.3Improved
Set alignment loss weight to 10.0 (ablation)CVT-SLR, CTC weight 1.0 · PHOENIX-2014WER: 20.2 -> 19.8; 20.3 -> 20.1Improved
Increase alignment loss weight from 10.0 to 100.0 or 1000.0 (ablation)CVT-SLR, CTC weight 1.0 · PHOENIX-2014WER: 19.8 -> 20.1 -> 20.5; 20.1 -> 20.6 -> 20.9Hurt
Use the full CVT-SLR method rather than the strongest single-cue baselinePHOENIX-2014 · PHOENIX-2014WER: 20.8 -> 19.8; 21.0 -> 20.1Improved
Use the full CVT-SLR method rather than the strongest multi-cue baselinePHOENIX-2014T · PHOENIX-2014TWER: 20.2 -> 19.4; 20.4 -> 20.3Mixed

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