hkchengrex / STCN

[NeurIPS 2021] Rethinking Space-Time Networks with Improved Memory Coverage for Efficient Video Object Segmentation

https://hkchengrex.com/STCN/

May I ask why this task can be called semi-supervised？

zhouweii234 opened this issue 2 years ago · comments

zhouweii234 commented 2 years ago

Why this task can be called semi-supervised, since we did not train the network with unlabeled data？

Rex Cheng commented 2 years ago

See DAVIS: https://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Perazzi_A_Benchmark_Dataset_CVPR_2016_paper.pdf. "Supervision" is in a per-video sense.