Posts tagged with "computer-vision"
- Hybrid UI Detection: Why We Split Vision and Intelligenceuitag combines Apple Vision's text detection with a fine-tuned YOLO model to hit 90.8% coverage on ScreenSpot-Pro — faster and cheaper than VLM-only approaches.
- Benchmarking UI Detection on ScreenSpot-ProHow we evaluated uitag against 1,581 annotations across 26 professional macOS applications — methodology, results, and what the numbers actually mean.