VideOCR v1.6.0

Highlights:

• Adds support for the new PaddleOCR v3.7.0 version which improved accuracy for Chinese, English, Japanese and Latin languages. Other languages also profit from the improved detection model.
• Adds option to save OCR detection/recognition images annotated with the detected boxes/text (currently limited to PaddleOCR), useful for reviewing or debugging OCR results
• Adds Drag and Drop Support (Currently limited to Windows due to an upstream issue on Linux)
• Adds option to disable frame stitching for better OCR accuracy
• Adds Copy CLI command to clipboard feature
• Adds option to reset settings back to default

Additional notes:

• Fixed a crash of the video preview of the GUI caused by a few select video files
• Fixed a mistake in the available PaddleOCR language list in the GUI
• Improved performance if a brightness threshold was applied
• Quite a few optimizations were done to shrink down the GUI and CLI executable size

Notes:

• Utilizes standalone version of PaddleOCR from https://github.com/timminator/PaddleOCR-Standalone
• Utilizes standalone version of Chrome-Lens-OCR from https://github.com/timminator/Chrome-Lens-OCR
• Linux version requires glibc 2.35 or newer (can be checked by running ldd --version in a terminal)

Download Tips:

• Windows:
  • CPU version:
    VideOCR-CPU-v1.6.0-setup-x64.exe
  • GPU version:
    VideOCR-GPU-v1.6.0-CUDA-11.8-setup-x64.exe (Nvidia 10 Series graphics cards)
    VideOCR-GPU-v1.6.0-CUDA-12.9-setup-x64.exe (Nvidia 16 - 50 Series graphics cards)
• Linux:
  • CPU version:
    VideOCR-CPU-v1.6.0-Linux.7z
  • GPU version:
    VideOCR-GPU-v1.6.0-CUDA-11.8-Linux.7z (Nvidia 10 Series graphics cards)
    VideOCR-GPU-v1.6.0-CUDA-12.9-Linux.7z.001 (Nvidia 16 - 50 Series graphics cards)
    VideOCR-GPU-v1.6.0-CUDA-12.9-Linux.7z.002 (Nvidia 16 - 50 Series graphics cards)