[CVPR26] CaptionFormer: Unified Segmentation, Tracking, and Captioning for Spatio-Temporal Objects

Gabriel Fiastre · Antoine Yang · Cordelia Schmid

Code Coming soon

Citation

If you use this work, please cite:

@inproceedings{fiastre2026captionformer,
  title     = {CaptionFormer: Learning to Jointly Segment and Caption Object Trajectories in Videos},
  author    = {Gabriel Fiastre and Antoine Yang and Cordelia Schmid},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year      = {2026}
}