Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →PaddleOCR text recognition skill using PP-OCRv5 lightweight model. Supports natural scene and complex document text detection and recognition. Use when user needs OCR text extraction from images.
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | -61% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 110% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 371% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 66% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -41% | 0% |
集成 SOTA 级轻量化 OCR 模型 PP-OCRv5，支持自然场景及复杂文档的文字检测与识别。
百度通用文字 OCR 识别，需设置 BAIDU_OCR_AK 和 BAIDU_OCR_SK。
bashpython3 scripts/baidu_ocr_text.py general /path/to/image.jpg python3 scripts/baidu_ocr_text.py accurate /path/to/image.jpg python3 scripts/baidu_ocr_text.py handwriting /path/to/note.jpg
Other measured skills in the registry, with their headline benchmark lift.