notesum.ai

Published at November 26

Learning Robust Anymodal Segmentor with Unimodal and Cross-modal Distillation

cs.CV

Released Date: November 26, 2024

Authors: Xu Zheng, Haiwei Xue, Jialei Chen, Yibo Yan, Lutao Jiang, Yuanhuiyi Lyu, Kailun Yang, Linfeng Zhang, Xuming Hu

Arxiv: http://arxiv.org/abs/2411.17141v1

Refer to caption

Method	Pub.	Training	Anymodal Evaluation							Mean
Method	Pub.	Training	F	E	L	FE	FL	EL	FEL	Mean
CMX (Zhang et al., 2023a)	T-ITS 2023		2.52	2.35	3.01	41.15	41.25	2.56	42.27	19.30
CMNeXt (Zhang et al., 2023b)	CVPR 2023	FEL	3.50	2.77	2.64	6.63	10.28	3.14	46.66	10.80
MAGIC (Zheng et al., 2024b)	ECCV 2024		43.22	2.68	22.95	43.51	49.05	22.98	49.02	33.34
Any2Seg (Zheng et al., 2024a)	ECCV 2024		44.40	3.17	22.33	44.51	49.96	22.63	50.00	33.86
Ours	-		46.01	19.57	32.13	46.29	51.25	35.21	51.14	40.23
w.r.t SoTA	-	-	+1.61	+16.40	+9.80	+1.78	+1.29	+12.58	+1.14	+6.37