Home News Latest Papers Tutorials Datasets Wiki SOTA LLM Models GPU Leaderboard Events

English

Image Captioning On Nocaps Val In Domain

Metrics

CIDEr

Pre-train (#images)

SPICE

Results

Performance results of various models on this benchmark

Model Name	CIDEr	Pre-train (#images)	SPICE	Paper Title	Repository
LEMON_base	107.7	200M	14.7	Scaling Up Vision-Language Pre-training for Image Captioning	-
BLIP-2 ViT-G OPT 6.7B (zero-shot)	123.7	1.1B	15.8	BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models
VinVL	103.1	5.7M	14.2	VinVL: Revisiting Visual Representations in Vision-Language Models
BLIP-2 ViT-G FlanT5 XL (zero-shot)	123.7	1.1B	16.3	BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models
BLIP_ViT-L	114.9	129M	15.2	BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation
SimVLM	113.7	1.8B	-	SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
LEMON_large	116.9	200M	15.8	Scaling Up Vision-Language Pre-training for Image Captioning	-
BLIP_CapFilt-L	111.8	129M	14.9	BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation
Enc-Dec	92.6	15M	12.5	Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual Concepts
OmniVL	104.6	14M	15	OmniVL:One Foundation Model for Image-Language and Video-Language Tasks	-
BLIP-2 ViT-G OPT 2.7B (zero-shot)	123	1.1B	15.8	BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models

0 of 11 row(s) selected.