Referring Expression Segmentation on RefCOCOg-val

Metric: Overall IoU (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

#	Model↕	Overall IoU▼	Extra Data	Paper	Date↕	Code
1	MLCD-Seg-7B	79.9	Yes	Multi-label Cluster Discrimination for Visual Re...	2024-07-24	Code
2	HyperSeg	79.4	Yes	HyperSeg: Towards Universal Visual Segmentation ...	2024-11-26	Code
3	UniLSeg-100	79.27	Yes	Universal Segmentation at Arbitrary Granularity ...	2023-12-04	Code
4	UniLSeg-20	78.41	Yes	Universal Segmentation at Arbitrary Granularity ...	2023-12-04	Code
5	EVF-SAM	78.2	Yes	EVF-SAM: Early Vision-Language Fusion for Text-P...	2024-06-28	Code
6	SegAgent	75.11	No	SegAgent: Exploring Pixel Understanding Capabili...	2025-03-11	Code
7	DETRIS	74.6	No	Densely Connected Parameter-Efficient Tuning for...	2025-01-15	Code
8	C3VG	74.43	No	Multi-task Visual Grounding with Coarse-to-Fine ...	2025-01-12	Code
9	GROUNDHOG	74.1	Yes	GROUNDHOG: Grounding Large Language Models to Ho...	2024-02-26	-
10	GLEE-Pro	72.9	Yes	General Object Foundation Model for Images and V...	2023-12-14	Code
11	SafaRi-B	70.48	Yes	SafaRi:Adaptive Sequence Transformer for Weakly ...	2024-07-02	-
12	PolyFormer-L	69.2	Yes	PolyFormer: Referring Image Segmentation as Sequ...	2023-02-14	Code
13	MaskRIS (Swin-B, combined DB)	69.12	No	MaskRIS: Semantic Distortion-aware Data Augmenta...	2024-11-28	Code
14	PolyFormer-B	67.76	Yes	PolyFormer: Referring Image Segmentation as Sequ...	2023-02-14	Code
15	MaskRIS (Swin-B)	65.55	No	MaskRIS: Semantic Distortion-aware Data Augmenta...	2024-11-28	Code
16	MagNet	65.36	No	Mask Grounding for Referring Image Segmentation	2023-12-19	Code
17	X-Decoder (Davit-d5)	64.6	Yes	Generalized Decoding for Pixel, Image, and Langu...	2022-12-21	Code
18	VLT (Swin-B)	63.49	No	VLT: Vision-Language Transformer and Query Gener...	2022-10-28	Code
19	LAVT	61.24	No	LAVT: Language-Aware Vision Transformer for Refe...	2021-12-04	Code
20	VLT (Darknet53)	52.99	No	Vision-Language Transformer and Query Generation...	2021-08-12	Code
21	SHNet	49.9	No	Comprehensive Multi-Modal Interactions for Refer...	2021-04-21	Code