HyperAI초신경

홈 뉴스 최신 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Zero Shot Semantic Segmentation On Coco Stuff

평가 지표

Inductive Setting hIoU

Transductive Setting hIoU

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	Inductive Setting hIoU	Transductive Setting hIoU	Paper Title	Repository
ZegCLIP	40.8	48.5	ZegCLIP: Towards Adapting CLIP for Zero-shot Semantic Segmentation
DeOP	38.2	-	Open-Vocabulary Semantic Segmentation with Decoupled One-Pass Network
CaGNet	18.2	19.5	Context-aware Feature Generation for Zero-shot Semantic Segmentation
STRICT	-	34.8	A Closer Look at Self-training for Zero-Label Semantic Segmentation
OTSeg	41.4	49.5	OTSeg: Multi-prompt Sinkhorn Attention for Zero-Shot Semantic Segmentation
SPNet	14.0	30.3	Semantic Projection Network for Zero- and Few-Label Semantic Segmentation
MaskCLIP+	-	45.0	Extract Free Dense Labels from CLIP
CLIP-RC	41.2	49.7	Exploring Regional Clues in CLIP for Zero-Shot Semantic Segmentation
MVP-SEG+	-	45.5	MVP-SEG: Multi-View Prompt Learning for Open-Vocabulary Semantic Segmentation	-
zsseg	36.3	41.5	A Simple Baseline for Open-Vocabulary Semantic Segmentation with Pre-trained Vision-language Model
ZegFormer	33.2	-	Decoupling Zero-Shot Semantic Segmentation
OTSeg+	41.5	49.8	OTSeg: Multi-prompt Sinkhorn Attention for Zero-Shot Semantic Segmentation
FreeSeg	-	45.3	FreeSeg: Free Mask from Interpretable Contrastive Language-Image Pretraining for Semantic Segmentation	-
SIGN	20.9	-	SIGN: Spatial-information Incorporated Generative Network for Generalized Zero-shot Semantic Segmentation	-
ZS5	15.0	16.2	Zero-Shot Semantic Segmentation

0 of 15 row(s) selected.