Papers with Code · CC BY-SA 4.0

Weakly-Supervised Semantic Segmentation

10-shot image generation › Semantic Segmentation › Weakly-Supervised Semantic Segmentation 182 평가 기록 10 datasets

The semantic segmentation task is to assign a label from a label set to each pixel in an image. In the case of fully supervised setting, the dataset consists of images and their corresponding pixel-level class-specific annotations (expensive pixel-level annotations). However, in the weakly-supervised setting, the dataset consists of images and corresponding annotations that are relatively easy to obtain, such as tags/labels of objects present in the image.

<span style="color:grey; opacity: 0.6">( Image credit: Weakly-Supervised Semantic Segmentation Network with Deep Seeded Region Growing )</span>

개요는 각 항목의 일부입니다. 위 탭에서 전체 기록을 페이지별로 탐색할 수 있습니다.

벤치마크

Papers with Code 아카이브의 데이터셋별 기록 예시입니다. 평가 조건과 지표 방향이 검증되지 않아 순위를 의미하지 않습니다.

Dataset기록된 모델 예시MetricPaperCode

PASCAL VOC 2012 val

SemPLeS (Swin-L)

Mean IoU: 83.4

논문 코드

PASCAL VOC 2012 test

SemPLeS (Swin-L)

Mean IoU: 82.9

논문 코드

COCO 2014 val

DHR (Swin-L, Mask2Former)

mIoU: 56.8

논문 코드

PASCAL VOC 2012 train

FMA-WSSS

Mean IoU: 80.4

논문 코드

ADE20K val

DHR (Swin-L, Mask2Former)

mIoU: 32.9

논문 코드

COCO-Stuff val

DHR (Swin-L, Mask2Former)

mIoU: 37.4

논문 코드

Cityscapes test

CARB

mIoU: 51.8

논문 코드

Cityscapes val

CARB

mIoU: 52.1

논문 코드

PASCAL Context val

DHR (Swin-L, Mask2Former)

mIoU: 53.6

논문 코드

^(#$!@#$)(()))******

Yair

0..5sec: Gg

논문 -

데이터셋

10개

PASCAL VOC 2012 val PASCAL VOC 2012 test COCO 2014 val PASCAL VOC 2012 train ADE20K val COCO-Stuff val Cityscapes test Cityscapes val PASCAL Context val ^(#$!@#$)(()))******

논문

Weakly Supervised Co-training with Swapping Assignments for Semantic Segmentation

2402.17891 2024-02-27 이 페이지의 평가 2건

mIoU: 53.7

WeakTr: Exploring Plain Vision Transformer for Weakly-supervised Semantic Segmentation

2304.01184 2023-04-03 이 페이지의 평가 2건

mIoU: 50.3

An Alternative to WSSS? An Empirical Study of the Segment Anything Model (SAM) on Weakly-Supervised Semantic Segmentation Problems

2305.01586 2023-05-02 이 페이지의 평가 1건

mIoU: 55.6

DHR: Dual Features-Driven Hierarchical Rebalancing in Inter- and Intra-Class Regions for Weakly-Supervised Semantic Segmentation

2404.00380 2024-03-30 이 페이지의 평가 1건

mIoU: 56.8

RecurSeed and EdgePredictMix: Pseudo-Label Refinement Learning for Weakly Supervised Semantic Segmentation across Single- and Multi-Stage Frameworks

2204.06754 2022-04-14 이 페이지의 평가 1건

mIoU: 46.4

Semantic Prompt Learning for Weakly-Supervised Semantic Segmentation

2401.11791 2024-01-22 이 페이지의 평가 1건

mIoU: 56.1

라이브러리

Papers with Code 리더보드 row에 연결된 구현 저장소입니다.

github.com/hustvl/weaktr 1 논문

github.com/youshyee/cosa 1 논문

github.com/Big-Brother-Pikachu/Text2Mask 1 논문

github.com/cenkbircanoglu/isim 1 논문

github.com/HAL-42/FMA-WSSS 1 논문

github.com/nvlabs/semples 1 논문

github.com/shjo-april/DHR 1 논문

github.com/shjo-april/mars 1 논문

Weakly-Supervised Semantic Segmentation

벤치마크

데이터셋

논문

Weakly Supervised Co-training with Swapping Assignments for Semantic Segmentation

WeakTr: Exploring Plain Vision Transformer for Weakly-supervised Semantic Segmentation

An Alternative to WSSS? An Empirical Study of the Segment Anything Model (SAM) on Weakly-Supervised Semantic Segmentation Problems

DHR: Dual Features-Driven Hierarchical Rebalancing in Inter- and Intra-Class Regions for Weakly-Supervised Semantic Segmentation

Fine-grained Background Representation for Weakly Supervised Semantic Segmentation

Foundation Model Assisted Weakly Supervised Semantic Segmentation

From Text to Mask: Localizing Entities Using the Attention of Text-to-Image Diffusion Models

MARS: Model-agnostic Biased Object Removal without Additional Supervision for Weakly-Supervised Semantic Segmentation

RecurSeed and EdgePredictMix: Pseudo-Label Refinement Learning for Weakly Supervised Semantic Segmentation across Single- and Multi-Stage Frameworks

Semantic Prompt Learning for Weakly-Supervised Semantic Segmentation

라이브러리