Metric: SPICE (higher is better)
| # | Model↕ | SPICE▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | GIT, Single Model | 15.7 | No | GIT: A Generative Image-to-text Transformer for ... | 2022-05-27 | Code |
| 2 | GIT2, Single Model | 15.62 | No | GIT: A Generative Image-to-text Transformer for ... | 2022-05-27 | Code |
| 3 | PaLI | 15.49 | No | PaLI: A Jointly-Scaled Multilingual Language-Ima... | 2022-09-14 | Code |
| 4 | CoCa - Google Brain | 15.13 | No | - | - | - |
| 5 | FudanFVL | 14.21 | No | - | - | - |
| 6 | Human | 14.21 | No | - | - | - |
| 7 | Single Model | 13.89 | No | SimVLM: Simple Visual Language Model Pretraining... | 2021-08-24 | Code |
| 8 | firethehole | 13.87 | No | - | - | - |
| 9 | FudanWYZ | 13.75 | No | - | - | - |
| 10 | Microsoft Cognitive Services team | 13.74 | No | VIVO: Visual Vocabulary Pre-Training for Novel O... | 2020-09-28 | - |
| 11 | IEDA-LAB | 12.52 | No | - | - | - |
| 12 | vll@mk514 | 12.14 | No | - | - | - |
| 13 | MD | 11.59 | No | - | - | - |
| 14 | ViTCAP-CIDEr-136.7-ENC-DEC-ViTbfocal10-test-CBS | 11.53 | No | - | - | - |
| 15 | VinVL (Microsoft Cognitive Services + MSR) | 11.48 | No | VinVL: Revisiting Visual Representations in Visi... | 2021-01-02 | Code |
| 16 | icgp2ssi1_coco_si_0.02_5_test | 11.43 | No | - | - | - |
| 17 | evertyhing | 11.18 | No | - | - | - |
| 18 | GRIT (zero-shot, no CBS, no VL pretraining, single model) | 11.1 | No | GRIT: Faster and Better Image captioning Transfo... | 2022-07-20 | Code |
| 19 | RCAL | 10.68 | No | - | - | - |
| 20 | vinvl_yuan_cbs | 10.57 | No | - | - | - |
| 21 | UpDown-C | 10.15 | No | - | - | - |
| 22 | Xinyi | 10.05 | No | - | - | - |
| 23 | cxy_nocaps_training | 10.01 | No | - | - | - |
| 24 | camel XE | 9.9 | No | - | - | - |
| 25 | UpDown + ELMo + CBS | 9.74 | No | - | - | - |
| 26 | Oscar | 9.72 | No | - | - | - |
| 27 | ClipCap (MLP + GPT2 tuning) | 9.7 | No | ClipCap: CLIP Prefix for Image Captioning | 2021-11-18 | Code |
| 28 | ClipCap (Transformer) | 9.57 | No | ClipCap: CLIP Prefix for Image Captioning | 2021-11-18 | Code |
| 29 | Check | 9.39 | No | - | - | - |
| 30 | 7_10-7_40000_predict_test.json | 9.35 | No | - | - | - |
| 31 | Neural Baby Talk + CBS | 8.77 | No | - | - | - |
| 32 | Neural Baby Talk | 8.2 | No | - | - | - |
| 33 | nocaps_training | 8.08 | No | - | - | - |
| 34 | UpDown | 8.08 | No | - | - | - |
| 35 | area_attention | 7.72 | No | - | - | - |
| 36 | Yu-Wu | 7.62 | No | - | - | - |
| 37 | B2 | 7.61 | No | - | - | - |
| 38 | YX | 7.52 | No | - | - | - |
| 39 | coco_all_19 | 7.4 | No | - | - | - |
| 40 | CS395T | 7.2 | No | - | - | - |