Image Captioning on COCO Captions

41개 결과 · ⬇ CSV · JSON

BLEU-4

Rank	Model	BLEU-4	CIDER	METEOR	SPICE	ROUGE-L	BLEU-1	BLEU-2	BLEU-3	Paper	Code	Year
1	mPLUG	46.5	155.1	32.0	26.0	–	–	–	–	mPLUG: Effective and Efficient Vision-Language Learning by Cross-modal Skip-connections	modelscope/modelscope · alibaba/AliceMind · x-plug/mplug	2022
2	OFA	44.9	154.9	32.5	26.6	–	–	–	–	OFA: Unifying Architectures, Tasks, and Modalities Through a Simple Sequence-to-Sequence Learning Framework	modelscope/modelscope · ofa-sys/ofa · JHKim-snu/GVCCI · +1	2022
3	GIT	44.1	151.1	32.2	26.3	–	–	–	–	GIT: A Generative Image-to-text Transformer for Vision and Language	microsoft/GenerativeImage2Text	2022
4	BLIP-2 ViT-G OPT 2.7B (zero-shot)	43.7	145.8	–	–	–	–	–	–	BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models	huggingface/transformers · salesforce/lavis · thudm/visualglm-6b · +14	2023
5	BLIP-2 ViT-G OPT 6.7B (zero-shot)	43.5	145.2	–	–	–	–	–	–	BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models	huggingface/transformers · salesforce/lavis · thudm/visualglm-6b · +14	2023
6	ExpansionNet v2 (No VL pretraining)	42.7	143.7	30.6	24.7	61.1	83.5	–	–	Exploiting Multiple Sequence Lengths in Fast End to End Training for Image Captioning	jchenghu/expansionnet_v2	2022
7	LEMON	42.6	145.5	31.4	25.5	–	–	–	–	Scaling Up Vision-Language Pre-training for Image Captioning		2021
8	BLIP-2 ViT-G FlanT5 XL (zero-shot)	42.4	144.5	–	–	–	–	–	–	BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models	huggingface/transformers · salesforce/lavis · thudm/visualglm-6b · +14	2023
8	GRIT (No VL pretraining - base)	42.4	144.2	30.6	24.3	60.7	84.2	–	–	GRIT: Faster and Better Image captioning Transformer Using Dual Visual Features	davidnvq/grit · pacman-ctm/thesis_code	2022
10	Prompt Tuning	41.81	141.4	31.51	24.42	–	–	–	–	Prompt Tuning for Generative Multimodal Pretrained Models	ofa-sys/ofa	2022
11	Oscar	41.7	140	30.6	24.5	–	–	–	–	Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks	rmokady/clip_prefix_caption · microsoft/Oscar · milvlg/rosita · +1	2020
12	Xmodal-Ctx	41.4	139.9	30.4	24.0	60.4	83.4	–	–	Beyond a Pre-Trained Object Detector: Cross-Modal Textual and Visual Context for Image Captioning	GT-RIPL/Xmodal-Ctx	2022
13	Xmodal-Ctx + OSCAR	41.3	142.2	–	24.9	–	–	–	–	Beyond a Pre-Trained Object Detector: Cross-Modal Textual and Visual Context for Image Captioning	GT-RIPL/Xmodal-Ctx	2022
13	X-VLM (base)	41.3	140.8	–	–	–	–	–	–	Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts	zengyan-97/x-vlm	2021
15	VinVL	41.0	140.9	31.1	25.2	–	–	–	–	VinVL: Revisiting Visual Representations in Vision-Language Models	microsoft/Oscar · pzzhang/VinVL · yaolinli/capenrich · +4	2021
16	CoCa	40.9	143.6	33.9	24.7	–	–	–	–	CoCa: Contrastive Captioners are Image-Text Foundation Models	mlfoundations/open_clip · facebookresearch/multimodal · lucidrains/CoCa-pytorch · +3	2022
17	SimVLM	40.6	143.3	33.4	25.4	–	–	–	–	SimVLM: Simple Visual Language Model Pretraining with Weak Supervision	yulong-XJTU/SimVLM · FerryHuang/SimVLM	2021
18	Prismer	40.4	136.5	31.4	24.4	–	–	–	–	Prismer: A Vision-Language Model with Multi-Task Experts	nvlabs/prismer · KastanDay/video-pretrained-transformer	2023
19	PTP-BLIP (14M)	40.1	135.0	30.4	23.7	–	–	–	–	Position-guided Text Prompt for Vision-Language Pre-training	sail-sg/ptp	2022
20	L-Verse	39.9	–	31.4	23.3	60.4	–	–	–	L-Verse: Bidirectional Generation Between Image and Text	tgisaturday/L-Verse	2021