Tokenizing Semantic Segmentation with RLE
Paper • 2602.21627 • Published
These are the trained models for our adaptation of Pix2Seq for token-based autoregressive video object detection and semantic segmentation.
Papers:
Code:
https://github.com/abhineet123/p2s-video
We also include all the pretrained models we have used in our experiments as well as a sample of evaluation data.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support