Fetching the paper…

SSTVOS: Sparse Spatiotemporal Transformers for Video Object Segmentation · Around