diff --git a/content/docs/about_model/model_summary.md b/content/docs/about_model/model_summary.md index 372df69f..07b7507f 100644 --- a/content/docs/about_model/model_summary.md +++ b/content/docs/about_model/model_summary.md @@ -37,9 +37,9 @@ katex: true | ch_PP-OCRv4_det_server_infer.onnx | 108M | 0.7922 | 0.8533 | 0.8216 | 3.9093 | | ch_ppocr_server_v2.0_det_infer.onnx | 47M | 0.7298 | 0.8128 | 0.7691 | 0.7419 | |||||| -| [damo/cv_proxylessnas_ocr-detection-db-line-level_damo](https://www.modelscope.cn/models/iic/cv_proxylessnas_ocr-detection-db-line-level_damo/summary) | 5.8M | 0.7666 | 0.8128 | 0.7890 | 0.6636 | -| [damo/cv_resnet18_ocr-detection-db-line-level_damo](https://www.modelscope.cn/models/iic/cv_resnet18_ocr-detection-db-line-level_damo/summary) | 47.2M | 0.7749 | 0.8167 | 0.7952 | 0.4121 | -| [damo/cv_resnet18_ocr-detection-line-level_damo](https://modelscope.cn/models/iic/cv_resnet18_ocr-detection-line-level_damo/summary) 未跑通 | 312M | - | - | - | - | +| [读光-文字检测-轻量化端侧DBNet行检测模型-中英-通用领域](https://www.modelscope.cn/models/iic/cv_proxylessnas_ocr-detection-db-line-level_damo/summary) | 5.8M | 0.7666 | 0.8128 | 0.7890 | 0.6636 | +| [读光-文字检测-DBNet行检测模型-中英-通用领域](https://www.modelscope.cn/models/iic/cv_resnet18_ocr-detection-db-line-level_damo/summary) | 47.2M | 0.7749 | 0.8167 | 0.7952 | 0.4121 | +| [读光-文字检测-行检测模型-中英-通用领域](https://modelscope.cn/models/iic/cv_resnet18_ocr-detection-line-level_damo/summary) 未跑通 | 312M | - | - | - | - | 不同推理引擎下,效果比较: @@ -63,6 +63,13 @@ katex: true |ch_PP-OCRv4_rec_server_infer.onnx | release/v2.7 | 86M | 0.7968 | 0.9381 | 0.6967 | |ch_PP-OCRv2_rec_infer.onnx | release/v2.3 | 8.0M | 0.6387 | 0.8398 | 0.6138| |ch_ppocr_mobile_v2.0_rec_infer.onnx | release/v2.0 | 4.3M | 0.5323 | 0.7823 | 0.5575| +|[读光-文字识别-行识别模型-中英-文档印刷体文本领域](https://www.modelscope.cn/models/iic/cv_convnextTiny_ocr-recognition-document_damo/summary) | - | 73M | 0.5968 | 0.7705 | - | +|[读光-文字识别-行识别模型-中英-通用领域](https://www.modelscope.cn/models/iic/cv_convnextTiny_ocr-recognition-general_damo/summary) | - | 73M | 0.5839 | 0.7615 | - | +|[读光-文字识别-行识别模型-中英-自然场景文本领域](https://www.modelscope.cn/models/iic/cv_convnextTiny_ocr-recognition-scene_damo/summary) | - | 73M | 0.5903 | 0.7779 | - | +|[读光-文字识别-轻量化端侧识别模型-中英-通用领域](https://www.modelscope.cn/models/iic/cv_LightweightEdge_ocr-recognitoin-general_damo/summary) | - | 7.4M | 0.5484 | 0.7515 | - | +|[读光-文字识别-CRNN模型-中英-通用领域](https://www.modelscope.cn/models/iic/cv_crnn_ocr-recognition-general_damo/summary) | - | 46M | 0.5935 | 0.7671 | - | +|[OFA文字识别-中文-通用场景-base](https://www.modelscope.cn/models/iic/ofa_ocr-recognition_general_base_zh/summary) 未跑通 | - | - | - | - | - | + 不同推理引擎下,效果比较: | 推理引擎 | 模型 | 模型大小 | Exact Match | Char Match | Speed(s/img) | @@ -74,7 +81,8 @@ katex: true - 输入Shape: - v2: `[3, 32, 320]` - v3~v4: `[3, 48, 320]` -- 不同模型,实例化如下: + +- 不同模型,实例化示例如下: ```python {linenos=table} from rapidocr_onnxruntime import RapidOCR