AI Paper Atlas

Research map / Architectures and efficiency / CNN design

CNN-N-Gram for Handwriting Word Recognition

Arik Poznanski, Lior Wolf

CVPR 2016 · Architectures and efficiency › CNN design · approach: Compact CNN architectures

This paper tried a common change: Add data augmentation

add test-side data augmentation · IAM · WER (%): 6.94 -> 6.45 Improved

What 689 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

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
use CNN-predicted n-gram attributes with CCA matchinghandwriting word recognition · IAMWER (%): 11.90 -> 6.45Improved
use CNN-predicted n-gram attributes with CCA matchinghandwriting word recognition · RIMESWER (%): 4.75 -> 3.90Improved
use CNN-predicted n-gram attributes with CCA matchinghandwriting word recognition · IFN/ENIT abcde-saccuracy (%): 88.50 -> 94.09Improved
use CNN-N-Gramprinted word recognition · SVT fullaccuracy (%): 80.7 -> 83.62Improved
use CNN-N-Gramprinted word recognition · IIIT5K-1000accuracy (%): 94.4 -> 94.15Hurt
use probability vectors instead of FC representations as CCA input (ablation)CNN-N-Gram · IAMWER (%): 6.45 -> 6.56No clear effect
remove trigrams during test (ablation)CNN-N-Gram · IAMWER (%): 6.45 -> 6.33No clear effect
remove bigrams and trigrams during test (ablation)CNN-N-Gram · IAMWER (%): 6.45 -> 6.29No clear effect
remove trigrams during training (ablation)CNN-N-Gram · IAMWER (%): 6.45 -> 6.32No clear effect
remove bigrams and trigrams during training (ablation)CNN-N-Gram · IAMWER (%): 6.45 -> 6.36No clear effect
use 19 specialized FC branches rather than 7 merged branches (ablation)CNN-N-Gram · IAMWER (%): 7.16 -> 6.45Improved
use 19 specialized FC branches rather than one FC branch (ablation)CNN-N-Gram · IAMWER (%): 7.81 -> 6.45Improved

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