Opt: use ONNX model converted from MXNet for faster OCR with MXNet fallback

deploy/Windows/config.py

@@ -38,6 +38,8 @@ class ConfigModel:
     StartOcrServer: bool = False
     OcrServerPort: int = 22268
     OcrClientAddress: str = "127.0.0.1:22268"
+    OcrBackend: str = "onnx"
+    IntraOpThreads: Optional[int] = None
 
     # Update
     EnableReload: bool = True