AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Measuring Human Perception to Improve Handwritten Document Transcription

Samuel Grieggs, Bingyu Shen, Greta Rauch, Pei Li, Jiaqi Ma, David Chiang, Brian L. Price, Walter J. Scheirer

TPAMI 2021 · Vision-language and multimodal › Human and embodied modeling · approach: Human perception modeling · 25 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 psychophysical loss (easy priority) to CTCCRNN · RIMES testCER (%): 7.47 -> 6.33Improved
add psychophysical loss (easy priority)CRNN · IAM testCER (%): 8.00 -> 7.76Improved
add psychophysical loss (easy priority)CTransformer · Latin testCER (%): 9.81 -> 8.96Improved
add psychophysical loss (easy priority)URNN · RIMES testCER (%): 6.75 -> 6.66No clear effect
add psychophysical loss (easy priority)CTransformer · Latin validationWER (%): 54.93 -> 55.39Hurt
prioritize hard examples instead of easy (ablation)psychophysical loss · IAM test (CRNN)CER (%): 7.76 -> 7.56Improved
prioritize hard examples instead of easy (ablation)psychophysical loss · CTransformer all datasetsCERHurt
replace reaction times with algorithmic hard example mining penalties (ablation)psychophysical loss · IAM test (CRNN)CER (%): 8.00 -> 7.76No clear effect
use reaction times for curriculum learning instead of loss (ablation)CRNN · IAM testCER (%): 8.00 -> 7.68Improved
use text length as penalty instead of reaction time (ablation)psychophysical loss · IAM test (CRNN)CER (%): 8.00 -> 7.92No clear effect
add psychophysical loss (H) with language model and synthetic pretrainingCRNN · IAM testCER (%): 4.71 -> 4.21Improved
reduce image height from 128 to 64 (ablation)CRNN · speedimages per second: 16 -> 31Mixed

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