AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Stochastic Fine-grained Labeling of Multi-state Sign Glosses for Continuous Sign Language Recognition

Zhe Niu, Brian Mak

ECCV 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 139 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 stochastic frame dropping (ablation)CSLR baseline · PHOENIX-2014 TestWER (%): 33.9 -> 30.3Improved
add stochastic frame dropping (ablation)CSLR baseline · PHOENIX-2014 DevWER (%): 34.1 -> 29.1Improved
add stochastic gradient stopping (pstop 75%) on top of SFD (ablation)SFD model · PHOENIX-2014 DevWER (%): 29.1 -> 29.8Hurt
add stochastic gradient stopping on top of SFD (ablation)SFD model · PHOENIX-2014 TestWER (%): 30.3 -> 29.4Improved
add deterministic fine-grained labeling with 2 states (ablation)SFD+SGS · PHOENIX-2014 DevWER (%): 29.8 -> 26.5Improved
add deterministic fine-grained labeling with 2 states (ablation)SFD+SGS · PHOENIX-2014 TestWER (%): 29.4 -> 27.2Improved
replace deterministic labeling with stochastic fine-grained labeling (Cmax 5) (ablation)SFD+SGS+DFL · PHOENIX-2014 DevWER (%): 26.5 -> 26.2Improved
replace deterministic labeling with stochastic fine-grained labeling (Cmax 5) (ablation)SFD+SGS+DFL · PHOENIX-2014 TestWER (%): 27.2 -> 26.8Improved
train state-number distribution with REINFORCE instead of uniform (ablation)SFL, Cmax=5 · PHOENIX-2014 DevWER (%): 30.2 -> 26.2Improved
train state-number distribution with REINFORCE instead of uniform (ablation)SFL, Cmax=5 · PHOENIX-2014 TestWER (%): 30.3 -> 26.8Improved
add the dataset language model during decodingSFD+SGS+SFL · PHOENIX-2014 DevWER (%): 26.2 -> 24.9Improved
add the dataset language model during decodingSFD+SGS+SFL · PHOENIX-2014 TestWER (%): 26.8 -> 25.3Improved

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