Deep spectral methods: a surprisingly strong baseline for unsupervised semantic segmentation and localization
Unsupervised localization and segmentation are long-standing computer vision challenges that involve decom-posing an image into semantically meaningful segments without any labeled data. These tasks are particularly interesting in an unsupervised setting due to the difficulty and cost of obtaining d...
Main Authors: | Melas-Kyriazi, L, Rupprecht, C, Laina, I, Vedaldi, A |
---|---|
格式: | Conference item |
語言: | English |
出版: |
IEEE
2022
|
相似書籍
-
Finding an unsupervised image segmenter in each of your deep generative models
由: Melas-Kyriazi, L, et al.
出版: (2021) -
RealFusion: 360 reconstruction of any object from a single image
由: Melas-Kyriazi, L, et al.
出版: (2023) -
Unsupervised multi-object segmentation by predicting probable motion patterns
由: Karazija, L, et al.
出版: (2022) -
Guess what moves: unsupervised video and image segmentation by anticipating motion
由: Choudhury, S, et al.
出版: (2022) -
Unsupervised part discovery from contrastive reconstruction
由: Choudhury, S, et al.
出版: (2021)