English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Advances in generative modeling based on GANs has motivated the community tofind their use beyond image generation and editing tasks. In particular,several recent works have shown that GAN representations can be re-purposed fordiscriminative tasks such as part segmentation, especially when training datais limited. But how do these improvements stack-up against recent advances inself-supervised learning? Motivated by this we present an alternative approachbased on contrastive learning and compare their performance on standardfew-shot part segmentation benchmarks. Our experiments reveal that not only dothe GAN-based approach offer no significant performance advantage, theirmulti-step training is complex, nearly an order-of-magnitude slower, and canintroduce additional bias. These experiments suggest that the inductive biasesof generative models, such as their ability to disentangle shape and texture,are well captured by standard feed-forward networks trained using contrastivelearning. These experiments suggest that the inductive biases present incurrent generative models, such as their ability to disentangle shape andtexture, are well captured by standard feed-forward networks trained usingcontrastive learning.

GANORCON: Are Generative Models Useful for Few-shot Segmentation?