The chinese-clip-opencv-onnxrun from hpc203

hpc203 / chinese-clip-opencv-onnxrun Goto Github PK

View Code? Open in Web Editor NEW

使用OpenCV+onnxruntime部署中文clip做以文搜图，给出一句话来描述想要的图片，就能从图库中搜出来符合要求的图片。包含C++和Python两个版本的程序

CMake 1.70% C++ 49.63% Python 48.67%

chinese-clip-opencv-onnxrun's People

Contributors

Stargazers

Watchers

chinese-clip-opencv-onnxrun's Issues

请问这里的转换后的预训练模型是哪个规模的？

如题。是 ViT-L/14 还是 ViT-B/16 还是 ViT-H/14，谢谢

执行C++版本的程序时，加载模型报错

您好，我在尝试使用C++版本的程序时，遇到了以下错误

[ERROR:0] global /home/wangweinan/prj/opencv-4.5.0/modules/dnn/src/onnx/onnx_importer.cpp (1786) handleNode DNN/ONNX: ERROR during processing node with 1 inputs and 1 outputs: [ReduceMean]:(/visual/ln_pre/ReduceMean_output_0)
terminate called after throwing an instance of 'cv::Exception'
  what():  OpenCV(4.5.0) /home/wangweinan/prj/opencv-4.5.0/modules/dnn/src/onnx/onnx_importer.cpp:1797: error: (-2:Unspecified error) in function 'handleNode'
> Node [ReduceMean]:(/visual/ln_pre/ReduceMean_output_0) parse error: OpenCV(4.5.0) /home/wangweinan/prj/opencv-4.5.0/modules/dnn/include/opencv2/dnn/dnn.inl.hpp:349: error: (-204:Requested object was not found) Required argument "keepdims" not found into dictionary in function 'get'
> 
已放弃 (核心已转储)

我使用自己转换的cn_clip模型也是有类似的错误

[ERROR:0] global /home/wangweinan/prj/opencv-4.5.0/modules/dnn/src/onnx/onnx_importer.cpp (1786) handleNode DNN/ONNX: ERROR during processing node with 2 inputs and 1 outputs: [Unsqueeze]:(/visual/Unsqueeze_output_0)
terminate called after throwing an instance of 'cv::Exception'
  what():  OpenCV(4.5.0) /home/wangweinan/prj/opencv-4.5.0/modules/dnn/src/onnx/onnx_importer.cpp:1797: error: (-2:Unspecified error) in function 'handleNode'
> Node [Unsqueeze]:(/visual/Unsqueeze_output_0) parse error: OpenCV(4.5.0) /home/wangweinan/prj/opencv-4.5.0/modules/dnn/src/onnx/onnx_importer.cpp:1309: error: (-215:Assertion failed) node_proto.input_size() == 1 in function 'handleNode'
>

请问这个问题有什么思路吗，如果不使用OpenCV有其他的C++ onnxruntime部署 cn_clip 的思路嘛