AI Paper Atlas

Research map / Video / Video segmentation and tracking

Self-Mutual Distillation Learning for Continuous Sign Language Recognition

Aiming Hao, Yuecong Min, Xilin Chen

ICCV 2021 · Video › Video segmentation and tracking · approach: Sign language and physiological video analysis · 167 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 weight normalization of the classifier (ablation)CTC baseline · PHOENIX14 testWER (%): 25.4 -> 24.7Improved
add weight normalization of the classifier (ablation)Baseline+WS · PHOENIX14 testWER (%): 21.9 -> 21.4Improved
add weight sharing between visual and contextual classifiers (ablation)CTC baseline · PHOENIX14 testWER (%): 24.7 -> 21.4Improved
add weight sharing between visual and contextual classifiers (ablation)CTC baseline · PHOENIX14 devWER (%): 24.1 -> 21.2Improved
update shared classifier with both LVFs and GCFs instead of only one (ablation)SMKD weight sharing · PHOENIX14 testWER (%): 22.5 -> 21.4Improved
add GSBA gloss segmentation to the visual module (ablation)CTC baseline · PHOENIX14 testWER (%): 24.7 -> 22.0Improved
use soft-target KD from contextual module instead of WS (ablation)CTC baseline · PHOENIX14 testWER (%): 21.4 -> 23.6Hurt
share classifier during synchronous training rather than independent classifiers (ablation)Baseline · PHOENIX14 testWER (%): 22.4 -> 21.4Improved
add gloss segmentation stage after sync training (ablation)Baseline+sync+dec train · PHOENIX14 testWER (%): 21.3 -> 21.0Improved
add decouple training stage at the end (ablation)Baseline+sync+gloss segment · PHOENIX14 testWER (%): 21.3 -> 21.0Improved
add decouple training stage after sync only (ablation)Baseline+sync · PHOENIX14 testWER (%): 21.4 -> 21.3No clear effect
vary alpha balancing visual and contextual CTC losses (ablation)SMKD weight sharing · PHOENIX14WER (%): optimal alpha 0.5Mixed

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