Image Captioning on nocaps entire

Metric: B3 (higher is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	B3▼	Extra Data	Paper	Date↕	Code
1	GIT, Single Model	57.68	No	GIT: A Generative Image-to-text Transformer for ...	2022-05-27	Code
2	CoCa - Google Brain	56.88	No	-	-	-
3	Microsoft Cognitive Services team	53.62	No	Scaling Up Vision-Language Pre-training for Imag...	2021-11-24	-
4	Prismer	52.48	No	Prismer: A Vision-Language Model with Multi-Task...	2023-03-04	Code
5	Single Model	51.06	No	SimVLM: Simple Visual Language Model Pretraining...	2021-08-24	Code
6	FudanFVL	50.84	No	-	-	-
7	FudanWYZ	49.58	No	-	-	-
8	IEDA-LAB	48.41	No	-	-	-
9	firethehole	48.14	No	-	-	-
10	MD	47.18	No	-	-	-
11	vll@mk514	46.13	No	-	-	-
12	ViTCAP-CIDEr-136.7-ENC-DEC-ViTbfocal10-test-CBS	45.26	No	-	-	-
13	VinVL (Microsoft Cognitive Services + MSR)	45.04	No	VinVL: Revisiting Visual Representations in Visi...	2021-01-02	Code
14	icgp2ssi1_coco_si_0.02_5_test	42.36	No	-	-	-
15	evertyhing	41.52	No	-	-	-
16	camel XE	40.68	No	-	-	-
17	vinvl_yuan_cbs	39.5	No	-	-	-
18	RCAL	39.11	No	-	-	-
19	Oscar	38.83	No	-	-	-
20	cxy_nocaps_training	37.56	No	-	-	-
21	Xinyi	37.39	No	-	-	-
22	MQ-UpDown-C	36.93	No	-	-	-
23	Human	36.37	No	-	-	-
24	UpDown + ELMo + CBS	35.39	No	-	-	-
25	nocaps_training	35.23	No	-	-	-
26	UpDown	35.23	No	-	-	-
27	B2	33.88	No	-	-	-
28	7_10-7_40000_predict_test.json	33.22	No	-	-	-
29	YX	31.74	No	-	-	-
30	None	31.7	No	-	-	-
31	area_attention	31.62	No	-	-	-
32	Neural Baby Talk	30.83	No	-	-	-
33	Neural Baby Talk + CBS	29.35	No	-	-	-
34	coco_all_19	28.64	No	-	-	-
35	Yu-Wu	25.76	No	-	-	-
36	CS395T	25.5	No	-	-	-

#1GIT, Single ModelSOTA
57.68
B3· 2022-05-27
GIT: A Generative Image-to-text Transformer for Vision and Language Code
#2CoCa - Google Brain
56.88
B3
No paper
#3Microsoft Cognitive Services teamSOTA
53.62
B3· 2021-11-24
Scaling Up Vision-Language Pre-training for Image Captioning
#4Prismer
52.48
B3· 2023-03-04
Prismer: A Vision-Language Model with Multi-Task Experts Code
#5Single ModelSOTA
51.06
B3· 2021-08-24
SimVLM: Simple Visual Language Model Pretraining with Weak Supervision Code
#6FudanFVL
50.84
B3
No paper
#7FudanWYZ
49.58
B3
No paper
#8IEDA-LAB
48.41
B3
No paper
#9firethehole
48.14
B3
No paper
#10MD
47.18
B3
No paper
#11vll@mk514
46.13
B3
No paper
#12ViTCAP-CIDEr-136.7-ENC-DEC-ViTbfocal10-test-CBS
45.26
B3
No paper
#13VinVL (Microsoft Cognitive Services + MSR)SOTA
45.04
B3· 2021-01-02
VinVL: Revisiting Visual Representations in Vision-Language Models Code
#14icgp2ssi1_coco_si_0.02_5_test
42.36
B3
No paper
#15evertyhing
41.52
B3
No paper
#16camel XE
40.68
B3
No paper
#17vinvl_yuan_cbs
39.5
B3
No paper
#18RCAL
39.11
B3
No paper
#19Oscar
38.83
B3
No paper
#20cxy_nocaps_training
37.56
B3
No paper
#21Xinyi
37.39
B3
No paper
#22MQ-UpDown-C
36.93
B3
No paper
#23Human
36.37
B3
No paper
#24UpDown + ELMo + CBS
35.39
B3
No paper
#25nocaps_training
35.23
B3
No paper
#26UpDown
35.23
B3
No paper
#27B2
33.88
B3
No paper
#287_10-7_40000_predict_test.json
33.22
B3
No paper
#29YX
31.74
B3
No paper
#30None
31.7
B3
No paper
#31area_attention
31.62
B3
No paper
#32Neural Baby Talk
30.83
B3
No paper
#33Neural Baby Talk + CBS
29.35
B3
No paper
#34coco_all_19
28.64
B3
No paper
#35Yu-Wu
25.76
B3
No paper
#36CS395T
25.5
B3
No paper