imageocr / main.py

Commit History

  1. Switch to Google Vision only - remove CLIP/transformers/torch

    3004157
  2. Improve CLIP accuracy: photo prompts, expanded categories, confidence threshold

    ca8026b
  3. Add Product Vision AI tab with Google Vision + CLIP fallback

    a8b8a13
  4. Add PDF support with per-page OCR extraction

    65b9e81
  5. Fix: lazy-load OCR model and smart URL auto-detection

    cbfd7c3
  6. Add Dockerfile and secure API Key for HuggingFace Spaces

    37a1c7e
  7. Fix Render port and python-bidi install error

    eec9584
  8. first commit

    a0a17f6