Image Retrieval on COCO (Common Objects in Context)

6개 결과 · ⬇ CSV · JSON

recall@1

Rank	Model	recall@1	recall@5	Recall@10	QPS	Paper	Code	Year
1	BLIP-2 ViT-G (fine-tuned)	68.3	87.7	92.6	–	BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models	huggingface/transformers · salesforce/lavis · thudm/visualglm-6b · +14	2023
2	VisualSparta	68.2	91.8	96.3	451.4	VisualSparta: An Embarrassingly Simple Approach to Large-scale Text-to-Image Search with Weighted Bag-of-words	soco-ai/SF-QA	2021
3	BLIP-2 ViT-L (fine-tuned)	66.3	86.5	91.8	–	BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models	huggingface/transformers · salesforce/lavis · thudm/visualglm-6b · +14	2023
4	FLAVA (zero-shot)	38.38	67.47	–	–	FLAVA: A Foundational Language And Vision Alignment Model	facebookresearch/multimodal · apsdehal/flava-tutorials · social-ai-studio/matk · +1	2021
5	CLIP (zero-shot)	33.29	62.47	–	–	FLAVA: A Foundational Language And Vision Alignment Model	facebookresearch/multimodal · apsdehal/flava-tutorials · social-ai-studio/matk · +1	2021
6	Oscar	–	–	98.3	–	Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks	rmokady/clip_prefix_caption · microsoft/Oscar · milvlg/rosita · +1	2020