AI Paper Atlas

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

Variational Connectionist Temporal Classification

Linlin Chao, Jingdong Chen, Wei Chu

ECCV 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning · 15 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
replace vanilla CTC with Var-CTCCRNN-CTC · IC03recall at 98% precision (%): 58.0 -> 90.3Improved
replace vanilla CTC with Var-CTCCRNN-CTC · IC13recall at 98% precision (%): 78.5 -> 92.1Improved
replace vanilla CTC with Var-CTCCRNN-CTC · IIIT5Krecall at 98% precision (%): 30.2 -> 66.4Improved
replace vanilla CTC with Var-CTCCRNN-CTC · SVTrecall at 98% precision (%): 34.3 -> 85.1Improved
add approximate posterior of blank on top of Mml-CTC (Var-CTC vs Mml-CTC) (ablation)Mml-CTC · SVTrecall at 98% precision (%): 57.7 -> 85.1Improved
add approximate posterior of blank on top of Mml-CTC (Var-CTC vs Mml-CTC) (ablation)Mml-CTC · IIIT5Krecall at 98% precision (%): 68.7 -> 66.4No clear effect
replace vanilla CTC with Var-CTCCRNN-CTC (ours) · IIIT5Ksequence accuracy (%): 79.4 -> 78.7No clear effect
replace vanilla CTC with Var-CTCCRNN-CTC (ours) · IC13sequence accuracy (%): 89.6 -> 88.7Hurt
replace vanilla CTC with Mml-CTCCRNN-CTC (ours) · IIIT5Ksequence accuracy (%): 79.4 -> 81.3Improved
replace vanilla CTC with Var-CTCResNet-CTC · IAM handwritingword error rate (%): 23.8 -> 24.0No clear effect
replace vanilla CTC with Var-CTCCTC · Seven scene text datasets (ASTER setup)recall at 98% precision (%): 67.57 -> 81.92Improved
replace vanilla CTC with Var-CTCCRNN-CTC · IIIT5KReplace error share (%): 49% -> 40%Improved

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