HyperAIHyperAI초신경
홈뉴스연구 논문튜토리얼데이터셋백과사전SOTALLM 모델GPU 랭킹컨퍼런스
전체 검색
소개
HyperAIHyperAI초신경

Image Captioning On Nocaps Val Out Domain

평가 지표

CIDEr
SPICE

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름
CIDEr
SPICE
Paper TitleRepository
Enc-Dec94.511.9Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual Concepts
BLIP_CapFilt-L111.514.2BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation
BLIP_ViT-L115.3 14.4BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation
BLIP-2 ViT-G FlanT5 XL (zero-shot)124.815.1BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models
OmniVL106.314.2OmniVL:One Foundation Model for Image-Language and Video-Language Tasks-
BLIP-2 ViT-G OPT 6.7B (zero-shot)124.414.8BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models
LEMON_large111.3 14.0Scaling Up Vision-Language Pre-training for Image Captioning-
BLIP-2 ViT-G OPT 2.7B (zero-shot)123.415.1BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models
SimVLM115.2-SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
VinVL 88.3 12.1VinVL: Revisiting Visual Representations in Vision-Language Models
0 of 10 row(s) selected.