AI Paper Atlas

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

Multi-Granularity Prediction for Scene Text Recognition

Peng Wang, Cheng Da, Cong Yao

ECCV 2022 · Vision-language and multimodal › Human and embodied modeling · approach: Language-grounded scene text and structure · 94 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
use 32x128 input with 4x4 patches instead of 224x224 with 16x16 (ablation)MGP-STR ViT · 6 STR benchmarks averageaverage accuracy: 91.07 -> 91.58Improved
add Character A3 module instead of using first T tokens (ablation)MGP-STR ViT · 6 STR benchmarks averageaverage accuracy: 91.58 -> 92.73Improved
use ViT-based vision model with A3 instead of CNN-based ABINet visionMGP-STR_Vision vs ABINet_V · 6 STR benchmarks averageaverage accuracy: 89.8 -> 92.73Improved
add Cumprod fusion of character and subword heads instead of character head only (ablation)MGP-STR_Fuse · 6 STR benchmarks averageaverage accuracy: 92.53 -> 93.35Improved
use Cumprod instead of Mean fusion (ablation)MGP-STR_Fuse · 6 STR benchmarks averageaverage accuracy: 93.28 -> 93.35No clear effect
add BPE head to character head (ablation)MGP-STR_C+B · 6 STR benchmarks averageaverage accuracy: 92.73 -> 93.24Improved
add WordPiece head to character head (ablation)MGP-STR_C+W · 6 STR benchmarks averageaverage accuracy: 92.73 -> 93.25Improved
add both BPE and WordPiece heads (ablation)MGP-STR_Fuse · 6 STR benchmarks averageaverage accuracy: 93.25 -> 93.35Improved
use subword head alone instead of character head (ablation)MGP-STR_Fuse · 6 STR benchmarks averageaverage accuracy: 92.53 -> 85.99Hurt
add BCN language model to vision model (ablation)MGP-STR_Vision · 6 STR benchmarks averageaverage accuracy: 92.40 -> 93.14Improved
use multi-granularity fusion instead of BCN explicit language modelMGP-STR_Fuse vs Vision+BCN · 6 STR benchmarks averageaverage accuracy: 93.14 -> 93.35Improved
add multi-granularity fusion on DeiT-Tiny (ablation)MGP-STR · 6 STR benchmarks average (DeiT-Tiny)average accuracy: 89.36 -> 89.91Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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