Papers with Code 아카이브

Unsupervised Semantic Segmentation with Language-image Pre-training

태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.

하위 태스크: 0
평가표: 12
보존된 평가 기록: 94

A segmentation task which does not utilise any human-level supervision for semantic segmentation except for a backbone which is initialised with features pre-trained with image-level labels.

코드

관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.

https://github.com/ExplainableML/cosmos

공식 코드

COSMOS: Cross-Modality Self-Distillation for Vision Language Pre-training

https://github.com/NVlabs/GroupViT

공식 코드

GroupViT: Semantic Segmentation Emerges from Text Supervision

https://github.com/NoelShin/reco

공식 코드

ReCo: Retrieve and Co-segment for Zero-shot Transfer

https://github.com/Qinying-Liu/TagAlign

공식 코드

TagAlign: Improving Vision-Language Alignment with Multi-Tag Classification

https://github.com/YuHengsss/Trident

공식 코드

Harnessing Vision Foundation Models for High-Performance, Training-Free Open Vocabulary Segmentation

https://github.com/avaxiao/TextRegion

공식 코드

TextRegion: Text-Aligned Region Tokens from Frozen Image-Text Models

https://github.com/chongzhou96/maskclip

공식 코드

Extract Free Dense Labels from CLIP

https://github.com/kakaobrain/tcl

공식 코드

Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs

https://github.com/linyq2117/tagclip

공식 코드

TagCLIP: A Local-to-Global Framework to Enhance Open-Vocabulary Multi-Label Classification of CLIP Without Training

https://github.com/m-arda-aydn/itaclip

공식 코드

ITACLIP: Boosting Training-Free Semantic Segmentation with Image, Text, and Architectural Enhancements

https://github.com/mc-lan/proxyclip

공식 코드

ProxyCLIP: Proxy Attention Improves CLIP for Open-Vocabulary Segmentation

https://github.com/shjo-april/TTD

공식 코드

TTD: Text-Tag Self-Distillation Enhancing Image-Text Alignment in CLIP to Alleviate Single Tag Bias

https://github.com/zdk258/CorrCLIP

공식 코드

CorrCLIP: Reconstructing Correlations in CLIP with Off-the-Shelf Foundation Models for Open-Vocabulary Semantic Segmentation

출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.