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

Existing Generalized Zero-shot Semantic Segmentation (GZLSS) methods applyeither finetuning the CLIP paradigm or formulating it as a mask classificationtask, benefiting from the Vision-Language Models (VLMs). However, thefine-tuning methods are restricted with fixed backbone models which are notflexible for segmentation, and mask classification methods heavily rely onadditional explicit mask proposers. Meanwhile, prevalent methods utilize onlyseen categories which is a great waste, i.e., neglecting the area exists butnot annotated. To this end, we propose CLIPTeacher, a new learning frameworkthat can be applied to various per-pixel classification segmentation modelswithout introducing any explicit mask proposer or changing the structure ofCLIP, and utilize both seen and ignoring areas. Specifically, CLIPTeacherconsists of two key modules: Global Learning Module (GLM) and Pixel LearningModule (PLM). Specifically, GLM aligns the dense features from an image encoderwith the CLS token, i.e., the only token trained in CLIP, which is a simple buteffective way to probe global information from the CLIP models. In contrast,PLM only leverages dense tokens from CLIP to produce high-level pseudoannotations for ignoring areas without introducing any extra mask proposer.Meanwhile, PLM can fully take advantage of the whole image based on the pseudoannotations. Experimental results on three benchmark datasets: PASCAL VOC 2012,COCO-Stuff 164k, and PASCAL Context show large performance gains, i.e., 2.2%,1.3%, and 8.8%

CLIP Is Also a Good Teacher: A New Learning Framework for Inductive Zero-shot Semantic Segmentation