Research map / Vision-language and multimodal / Human and embodied modeling
CDistNet: Perceiving Multi-domain Character Distance for Robust Text Recognition
Tianlun Zheng, Zhineng Chen, Shancheng Fang, Hongtao Xie, Yu–Gang Jiang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add semantic branch (position queries semantic feature) (ablation) | CDistNet · IC15 | accuracy: 82.11 -> 85.92 | Improved |
| add semantic branch (ablation) | CDistNet · SVT | accuracy: 90.26 -> 93.66 | Improved |
| replace Add fusion with dynamic shared fusion (DSF) (ablation) | CDistNet · SVT | accuracy: 91.96 -> 93.66 | Improved |
| replace Dot fusion with DSF (ablation) | CDistNet · SVT | accuracy: 91.81 -> 93.66 | Improved |
| share weights among MDCDP modules (ablation) | CDistNet · SVT | accuracy: 93.51 -> 93.66 | Improved |
| apply SAE to position branch only instead of none (ablation) | CDistNet · SVT | accuracy: 92.27 -> 93.66 | Improved |
| apply SAE to all three branches instead of position only (ablation) | CDistNet · IC15 | accuracy: 85.92 -> 84.70 | Hurt |
| use 3 MDCDP modules instead of 1 (ablation) | CDistNet · IC15 | accuracy: 84.82 -> 85.92 | Improved |
| use 4 MDCDP modules instead of 3 (ablation) | CDistNet · IC15 | accuracy: 85.92 -> 84.43 | Hurt |
| use 2 MDCDP modules instead of 3 (ablation) | CDistNet · SVT | accuracy: 93.66 -> 94.28 | Mixed |
| remove TPS rectification block (ablation) | CDistNet · CUTE80 | accuracy: 93.4 -> 92.0 | Hurt |
| use position attention on semantic feature (P_S) instead of semantic attention S_V (ablation) | CDistNet · SVT | accuracy: 91.34 -> 93.66 | Improved |
Closest papers by meaning
- OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoning (NeurIPS 2025 · similarity 0.89)
- ASCIIEval: Benchmarking Models' Visual Perception in Text Strings via ASCII Art (ICLR 2026 · similarity 0.88)
- Boosting Semi-Supervised Scene Text Recognition via Viewing and Summarizing (NeurIPS 2024 · similarity 0.87)
- OTE: Exploring Accurate Scene Text Recognition Using One Token (CVPR 2024 · similarity 0.87)
- T-MARS: Improving Visual Representations by Circumventing Text Feature Learning (ICLR 2024 · similarity 0.87)
- AlignVLM: Bridging Vision and Language Latent Spaces for Multimodal Document Understanding (NeurIPS 2025 · similarity 0.86)
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.
