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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Although recent advancements in diffusion models enabled high-fidelity anddiverse image generation, training of discriminative models largely depends oncollections of massive real images and their manual annotation. Here, wepresent a training method for semantic segmentation that neither relies on realimages nor manual annotation. The proposed method {\it attn2mask} utilizesimages generated by a text-to-image diffusion model in combination with itsinternal text-to-image cross-attention as supervisory pseudo-masks. Since thetext-to-image generator is trained with image-caption pairs but withoutpixel-wise labels, attn2mask can be regarded as a weakly supervisedsegmentation method overall. Experiments show that attn2mask achieves promisingresults in PASCAL VOC for not using real training data for segmentation at all,and it is also useful to scale up segmentation to a more-class scenario, i.e.,ImageNet segmentation. It also shows adaptation ability with LoRA-basedfine-tuning, which enables the transfer to a distant domain i.e., Cityscapes.

Attention as Annotation: Generating Images and Pseudo-masks for Weakly Supervised Semantic Segmentation with Diffusion