Timezone: »

Object Segmentation Without Labels with Large-Scale Generative Models
Andrey Voynov · Stanislav Morozov · Artem Babenko

Thu Jul 22 09:00 PM -- 11:00 PM (PDT) @

The recent rise of unsupervised and self-supervised learning has dramatically reduced the dependency on labeled data, providing high-quality representations for transfer on downstream tasks. Furthermore, recent works also employed these representations in a fully unsupervised setup for image classification, reducing the need for human labels on the fine-tuning stage as well. This work demonstrates that large-scale unsupervised models can also perform a more challenging object segmentation task, requiring neither pixel-level nor image-level labeling. Namely, we show that recent unsupervised GANs allow to differentiate between foreground/background pixels, providing high-quality saliency masks. By extensive comparison on common benchmarks, we outperform existing unsupervised alternatives for object segmentation, achieving new state-of-the-art.

Author Information

Andrey Voynov (Yandex)
Stanislav Morozov (Yandex)
Artem Babenko (Yandex)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors