Optical Character Recognition Ocr On Videodb
评估指标
Average Accuracy
Character Error Rate (CER)
Word Error Rate (WER)
评测结果
各个模型在此基准测试上的表现结果
| Paper Title | Repository | ||||
|---|---|---|---|---|---|
| GPT-4o | 76.22 | 0.2378 | 0.5117 | Benchmarking Vision-Language Models on Optical Character Recognition in Dynamic Video Environments | |
| Gemini-1.5 Pro | 76.13 | 0.2387 | 0.2385 | Benchmarking Vision-Language Models on Optical Character Recognition in Dynamic Video Environments | |
| Claude-3 Sonnet | 67.71 | 0.3229 | 0.4663 | Benchmarking Vision-Language Models on Optical Character Recognition in Dynamic Video Environments | |
| RapidOCR | 56.98 | 0.7620 | 0.4302 | Benchmarking Vision-Language Models on Optical Character Recognition in Dynamic Video Environments | |
| EasyOCR | 49.30 | 0.5070 | 0.8262 | Benchmarking Vision-Language Models on Optical Character Recognition in Dynamic Video Environments |
0 of 5 row(s) selected.