Metric: SPICE (higher is better)
| # | Model↕ | SPICE▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | GIT2, Single Model | 16.11 | No | GIT: A Generative Image-to-text Transformer for ... | 2022-05-27 | Code |
| 2 | GIT, Single Model | 15.96 | No | GIT: A Generative Image-to-text Transformer for ... | 2022-05-27 | Code |
| 3 | PaLI | 15.75 | No | PaLI: A Jointly-Scaled Multilingual Language-Ima... | 2022-09-14 | Code |
| 4 | PaLI | 15.75 | No | PaLI: A Jointly-Scaled Multilingual Language-Ima... | 2022-09-14 | Code |
| 5 | CoCa - Google Brain | 15.54 | No | - | - | - |
| 6 | Microsoft Cognitive Services team | 15.06 | No | VIVO: Visual Vocabulary Pre-Training for Novel O... | 2020-09-28 | - |
| 7 | firethehole | 14.88 | No | - | - | - |
| 8 | FudanFVL | 14.79 | No | - | - | - |
| 9 | Human | 14.72 | No | - | - | - |
| 10 | FudanWYZ | 14.71 | No | - | - | - |
| 11 | Single Model | 14.61 | No | SimVLM: Simple Visual Language Model Pretraining... | 2021-08-24 | Code |
| 12 | vll@mk514 | 14.37 | No | - | - | - |
| 13 | IEDA-LAB | 14.15 | No | - | - | - |
| 14 | MD | 13.64 | No | - | - | - |
| 15 | VinVL (Microsoft Cognitive Services + MSR) | 13.36 | No | VinVL: Revisiting Visual Representations in Visi... | 2021-01-02 | Code |
| 16 | ViTCAP-CIDEr-136.7-ENC-DEC-ViTbfocal10-test-CBS | 12.98 | No | - | - | - |
| 17 | RCAL | 12.47 | No | - | - | - |
| 18 | evertyhing | 12.24 | No | - | - | - |
| 19 | camel XE | 12.14 | No | - | - | - |
| 20 | vinvl_yuan_cbs | 12.12 | No | - | - | - |
| 21 | icgp2ssi1_coco_si_0.02_5_test | 12.11 | No | - | - | - |
| 22 | Xinyi | 11.88 | No | - | - | - |
| 23 | MQ-UpDown-C | 11.87 | No | - | - | - |
| 24 | cxy_nocaps_training | 11.81 | No | - | - | - |
| 25 | Oscar | 11.53 | No | - | - | - |
| 26 | UpDown + ELMo + CBS | 11.45 | No | - | - | - |
| 27 | ClipCap (MLP + GPT2 tuning) | 11.26 | No | ClipCap: CLIP Prefix for Image Captioning | 2021-11-18 | Code |
| 28 | 7_10-7_40000_predict_test.json | 11.14 | No | - | - | - |
| 29 | ClipCap (Transformer) | 10.92 | No | ClipCap: CLIP Prefix for Image Captioning | 2021-11-18 | Code |
| 30 | nocaps_training | 10.33 | No | - | - | - |
| 31 | UpDown | 10.33 | No | - | - | - |
| 32 | None | 10.28 | No | - | - | - |
| 33 | Neural Baby Talk + CBS | 9.83 | No | - | - | - |
| 34 | YX | 9.7 | No | - | - | - |
| 35 | area_attention | 9.7 | No | - | - | - |
| 36 | B2 | 9.54 | No | - | - | - |
| 37 | coco_all_19 | 9.28 | No | - | - | - |
| 38 | Neural Baby Talk | 9.26 | No | - | - | - |
| 39 | Yu-Wu | 8.37 | No | - | - | - |
| 40 | CS395T | 8.28 | No | - | - | - |