Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387389614> ?p ?o ?g. }
Showing items 1 to 67 of
67
with 100 items per page.
- W4387389614 abstract "Existing Generalized Zero-shot Semantic Segmentation (GZLSS) methods apply either finetuning the CLIP paradigm or formulating it as a mask classification task, benefiting from the Vision-Language Models (VLMs). However, the fine-tuning methods are restricted with fixed backbone models which are not flexible for segmentation, and mask classification methods heavily rely on additional explicit mask proposers. Meanwhile, prevalent methods utilize only seen categories which is a great waste, i.e., neglecting the area exists but not annotated. To this end, we propose CLIPTeacher, a new learning framework that can be applied to various per-pixel classification segmentation models without introducing any explicit mask proposer or changing the structure of CLIP, and utilize both seen and ignoring areas. Specifically, CLIPTeacher consists of two key modules: Global Learning Module (GLM) and Pixel Learning Module (PLM). Specifically, GLM aligns the dense features from an image encoder with the CLS token, i.e., the only token trained in CLIP, which is a simple but effective way to probe global information from the CLIP models. In contrast, PLM only leverages dense tokens from CLIP to produce high-level pseudo annotations for ignoring areas without introducing any extra mask proposer. Meanwhile, PLM can fully take advantage of the whole image based on the pseudo annotations. 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%" @default.
- W4387389614 created "2023-10-06" @default.
- W4387389614 creator A5029585911 @default.
- W4387389614 creator A5047964969 @default.
- W4387389614 creator A5049129547 @default.
- W4387389614 creator A5075576624 @default.
- W4387389614 creator A5085309296 @default.
- W4387389614 date "2023-10-03" @default.
- W4387389614 modified "2023-10-07" @default.
- W4387389614 title "CLIP Is Also a Good Teacher: A New Learning Framework for Inductive Zero-shot Semantic Segmentation" @default.
- W4387389614 doi "https://doi.org/10.48550/arxiv.2310.02296" @default.
- W4387389614 hasPublicationYear "2023" @default.
- W4387389614 type Work @default.
- W4387389614 citedByCount "0" @default.
- W4387389614 crossrefType "posted-content" @default.
- W4387389614 hasAuthorship W4387389614A5029585911 @default.
- W4387389614 hasAuthorship W4387389614A5047964969 @default.
- W4387389614 hasAuthorship W4387389614A5049129547 @default.
- W4387389614 hasAuthorship W4387389614A5075576624 @default.
- W4387389614 hasAuthorship W4387389614A5085309296 @default.
- W4387389614 hasBestOaLocation W43873896141 @default.
- W4387389614 hasConcept C111919701 @default.
- W4387389614 hasConcept C118505674 @default.
- W4387389614 hasConcept C119857082 @default.
- W4387389614 hasConcept C13280743 @default.
- W4387389614 hasConcept C154945302 @default.
- W4387389614 hasConcept C160633673 @default.
- W4387389614 hasConcept C185798385 @default.
- W4387389614 hasConcept C199360897 @default.
- W4387389614 hasConcept C204321447 @default.
- W4387389614 hasConcept C205649164 @default.
- W4387389614 hasConcept C38652104 @default.
- W4387389614 hasConcept C41008148 @default.
- W4387389614 hasConcept C48145219 @default.
- W4387389614 hasConcept C75608658 @default.
- W4387389614 hasConcept C89600930 @default.
- W4387389614 hasConceptScore W4387389614C111919701 @default.
- W4387389614 hasConceptScore W4387389614C118505674 @default.
- W4387389614 hasConceptScore W4387389614C119857082 @default.
- W4387389614 hasConceptScore W4387389614C13280743 @default.
- W4387389614 hasConceptScore W4387389614C154945302 @default.
- W4387389614 hasConceptScore W4387389614C160633673 @default.
- W4387389614 hasConceptScore W4387389614C185798385 @default.
- W4387389614 hasConceptScore W4387389614C199360897 @default.
- W4387389614 hasConceptScore W4387389614C204321447 @default.
- W4387389614 hasConceptScore W4387389614C205649164 @default.
- W4387389614 hasConceptScore W4387389614C38652104 @default.
- W4387389614 hasConceptScore W4387389614C41008148 @default.
- W4387389614 hasConceptScore W4387389614C48145219 @default.
- W4387389614 hasConceptScore W4387389614C75608658 @default.
- W4387389614 hasConceptScore W4387389614C89600930 @default.
- W4387389614 hasLocation W43873896141 @default.
- W4387389614 hasOpenAccess W4387389614 @default.
- W4387389614 hasPrimaryLocation W43873896141 @default.
- W4387389614 hasRelatedWork W1663079876 @default.
- W4387389614 hasRelatedWork W2069133146 @default.
- W4387389614 hasRelatedWork W2130974462 @default.
- W4387389614 hasRelatedWork W2378211422 @default.
- W4387389614 hasRelatedWork W2967478618 @default.
- W4387389614 hasRelatedWork W2970530566 @default.
- W4387389614 hasRelatedWork W2997152889 @default.
- W4387389614 hasRelatedWork W4321353415 @default.
- W4387389614 hasRelatedWork W4385572700 @default.
- W4387389614 hasRelatedWork W972276598 @default.
- W4387389614 isParatext "false" @default.
- W4387389614 isRetracted "false" @default.
- W4387389614 workType "article" @default.