Document Image Classification
Document image classification is the task of classifying documents based on images of their contents.
( Image credit: Real-Time Document Image Classification using Deep CNN and Extreme Learning Machines )
Papers
Showing 1–10 of 50 papers
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | EAML | Accuracy | 97.7 | — | Unverified |
| 2 | Cross-Modal | Accuracy | 97.05 | — | Unverified |
| 3 | DocFormerBASE | Accuracy | 96.17 | — | Unverified |
| 4 | LayoutLMV3Large | Accuracy | 95.93 | — | Unverified |
| 5 | LiLT[EN-R]BASE | Accuracy | 95.68 | — | Unverified |
| 6 | LayoutLMv2LARGE | Accuracy | 95.64 | — | Unverified |
| 7 | TILT-Large | Accuracy | 95.52 | — | Unverified |
| 8 | DocFormer large | Accuracy | 95.5 | — | Unverified |
| 9 | LayoutLMv3BASE | Accuracy | 95.44 | — | Unverified |
| 10 | Donut | Accuracy | 95.3 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | DocXClassifier-L | Accuracy | 95.57 | — | Unverified |
| 2 | DocBert [DOCBERT] | Accuracy | 91.95 | — | Unverified |
| 3 | Eff-GNN + Word2Vec [word2vec] | Accuracy | 91 | — | Unverified |
| 4 | Multimodal Side-Tuning (MobileNetV2) | Accuracy | 90.5 | — | Unverified |
| 5 | Multimodal Side-Tuning (ResNet50) | Accuracy | 90.3 | — | Unverified |
| 6 | DocBERT [DOCBERT] | Accuracy | 82.3 | — | Unverified |
| 7 | BERT [BERT] | Accuracy | 79 | — | Unverified |
| 8 | Eff-GNN + Word2Vec [word2vec] + Image Embedding | Accuracy | 77.5 | — | Unverified |
| 9 | Eff-GNN+ Word2Vec [word2vec] | Accuracy | 73.5 | — | Unverified |
| 10 | VGG | Memory | 7.08 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | PCGAN-CHAR | Accuracy | 89.54 | — | Unverified |
| 2 | Pixel-level RC | Accuracy | 77.22 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | PCGAN-CHAR | Accuracy | 96.68 | — | Unverified |
| 2 | Pixel-level RC | Accuracy | 95.46 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | ResNet-RS (ResNet-200 + RS training tricks) | Top 1 Accuracy - Verb | 83.4 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Pixel-level RC | Accuracy | 97.62 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | PCGAN-CHAR | Accuracy | 98.43 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | CNN | Accuracy | 86 | — | Unverified |