Generalized Referring Expression Segmentation

Metrics

cIoU

gIoU

Results

Performance results of various models on this benchmark

			Paper Title
GSVA-Llama2-13B	66.38	70.04	GSVA: Generalized Segmentation via Multimodal Large Language Models
MABP	65.69	68.79	Bring Adaptive Binding Prototypes to Generalized Referring Expression Segmentation
HDC	65.42	68.28	CoHD: A Counting-Aware Hierarchical Decoding Framework for Generalized Referring Expression Segmentation
GSVA-Vicuna-13B-v1.1	64.05	68.01	GSVA: Generalized Segmentation via Multimodal Large Language Models
GROUNDHOG	-	66.70	GROUNDHOG: Grounding Large Language Models to Holistic Segmentation
GSVA-Vicuna-7B-v1.1	63.29	66.47	GSVA: Generalized Segmentation via Multimodal Large Language Models
ReLA	62.42	63.60	GRES: Generalized Referring Expression Segmentation
LAVT	57.64	58.40	LAVT: Language-Aware Vision Transformer for Referring Image Segmentation
CRIS	55.34	56.27	CRIS: CLIP-Driven Referring Image Segmentation
LTS	52.30	52.70	Locate then Segment: A Strong Pipeline for Referring Image Segmentation
VLT	52.51	52.00	Vision-Language Transformer and Query Generation for Referring Segmentation
MattNet	47.51	48.24	MAttNet: Modular Attention Network for Referring Expression Comprehension

0 of 12 row(s) selected.

Generalized Referring Expression Segmentation

Metrics

cIoU

gIoU

Results

Performance results of various models on this benchmark

			Paper Title
GSVA-Llama2-13B	66.38	70.04	GSVA: Generalized Segmentation via Multimodal Large Language Models
MABP	65.69	68.79	Bring Adaptive Binding Prototypes to Generalized Referring Expression Segmentation
HDC	65.42	68.28	CoHD: A Counting-Aware Hierarchical Decoding Framework for Generalized Referring Expression Segmentation
GSVA-Vicuna-13B-v1.1	64.05	68.01	GSVA: Generalized Segmentation via Multimodal Large Language Models
GROUNDHOG	-	66.70	GROUNDHOG: Grounding Large Language Models to Holistic Segmentation
GSVA-Vicuna-7B-v1.1	63.29	66.47	GSVA: Generalized Segmentation via Multimodal Large Language Models
ReLA	62.42	63.60	GRES: Generalized Referring Expression Segmentation
LAVT	57.64	58.40	LAVT: Language-Aware Vision Transformer for Referring Image Segmentation
CRIS	55.34	56.27	CRIS: CLIP-Driven Referring Image Segmentation
LTS	52.30	52.70	Locate then Segment: A Strong Pipeline for Referring Image Segmentation
VLT	52.51	52.00	Vision-Language Transformer and Query Generation for Referring Segmentation
MattNet	47.51	48.24	MAttNet: Modular Attention Network for Referring Expression Comprehension

0 of 12 row(s) selected.

Generalized Referring Expression Segmentation | SOTA | HyperAI