AI Paper Atlas

Research map / Video / Video segmentation and tracking

Visual Alignment Constraint for Continuous Sign Language Recognition

Yuecong Min, Aiming Hao, Xiujuan Chai, Xilin Chen

ICCV 2021 · Video › Video segmentation and tracking · approach: Sign language and physiological video analysis · 232 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 visual enhancement (VE) loss (ablation)CTC baseline CSLR · PHOENIX14 DevWER (%): 25.4 -> 23.3Improved
add visual alignment (VA) distillation loss (ablation)CTC baseline CSLR · PHOENIX14 DevWER (%): 25.4 -> 24.5Improved
add both VE and VA losses (VAC) (ablation)CTC baseline CSLR · PHOENIX14 DevWER (%): 25.4 -> 21.2Improved
add both VE and VA losses (VAC) (ablation)CTC baseline CSLR · PHOENIX14 TestWER (%): 26.6 -> 22.3Improved
add VA loss on top of VE loss (ablation)VE-only model · PHOENIX14 DevWER (%): 23.3 -> 21.2Improved
add BN after each temporal convolution (iteration 1) (ablation)CSLR training · PHOENIX14 DevWER (%): 32.7 -> 27.2Improved
add BN after each temporal convolution (non-iterative) (ablation)CSLR training · PHOENIX14 DevWER (%): 30.4 -> 25.4Improved
add iterative training (3 iterations) to BN baseline (ablation)CSLR training · PHOENIX14 DevWER (%): 25.4 -> 24.7Improved
lower feature-extractor learning rate to 0.1 of alignment model (ablation)CSLR training · PHOENIX14 DevWER (%): 25.4 -> 25.0Improved
raise feature-extractor learning rate to 10x alignment model (ablation)CSLR training · PHOENIX14 DevWER (%): 25.4 -> 34.8Hurt
use VA temperature 8 instead of 1 (ablation)VAC · PHOENIX14 DevWER (%): 22.1 -> 21.2Improved
raise VA loss weight from 25 to 35 (ablation)VAC · PHOENIX14 DevWER (%): 21.2 -> 22.0Hurt

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