HyperAI초신경

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

한국어

HyperAI초신경

Interactive Segmentation On Davis

평가 지표

NoC@85

NoC@90

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	NoC@85	NoC@90	Paper Title	Repository
FocalClick-B3-S2	2.92	4.52	FocalClick: Towards Practical Interactive Image Segmentation
RITM (HRNet18, SBD)	4.36	5.74	Reviving Iterative Training with Mask Guidance for Interactive Segmentation
BRS	5.58	8.24	Interactive Image Segmentation via Backpropagating Refinement Scheme	-
RITM (HRNet-32, C+L)	4.11	5.34	Reviving Iterative Training with Mask Guidance for Interactive Segmentation
ICL CFR-1 (ViT-H, C+L)	3	4.24	CFR-ICL: Cascade-Forward Refinement with Iterative Click Loss for Interactive Image Segmentation
Latent diversity	5.05	9.57	Interactive Image Segmentation With Latent Diversity	-
IA+SA	5.16	-	Continuous Adaptation for Interactive Object Segmentation by Learning from Corrections	-
DOS with GC	9.03	12.58	Deep Interactive Object Selection
ViT-B+MST+CL	-	4.55	MST: Adaptive Multi-Scale Tokens Guided Interactive Segmentation
f-BRS-B (ResNet-101)	5.04	7.41	f-BRS: Rethinking Backpropagating Refinement for Interactive Segmentation
DOS w/o GC	12.52	17.11	Deep Interactive Object Selection
IA-FP-Net	4.03	5.22	Cascaded Sparse Feature Propagation Network for Interactive Segmentation
SimpleClick (ViT-H, SBD)	4.20	5.34	SimpleClick: Interactive Image Segmentation with Simple Vision Transformers
SimpleClick (ViT-H, C+L)	3.41	4.70	SimpleClick: Interactive Image Segmentation with Simple Vision Transformers
EdgeFlow	4.54	5.77	EdgeFlow: Achieving Practical Interactive Segmentation with Edge-Guided Flow

0 of 15 row(s) selected.