https://github.com/OpenGVLab/all-seeing

[flrngel]

Summary

All-Seeing 1B (AS-1B) dataset: we propose a new large-scale dataset (AS-1B) for open-world panoptic visual recognition and understanding, using an economical semi-automatic data engine that combines the power of off-the-shelf vision/language models and human feedback.

Tests

Memo

• Slow for production
• Model performs OCR, without specifically training the task.