Metric: CIDEr (higher is better)
| # | Model↕ | CIDEr▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | PaLI | 126.67 | No | PaLI: A Jointly-Scaled Multilingual Language-Ima... | 2022-09-14 | Code |
| 2 | GIT2, Single Model | 122.27 | No | GIT: A Generative Image-to-text Transformer for ... | 2022-05-27 | Code |
| 3 | GIT, Single Model | 122.04 | No | GIT: A Generative Image-to-text Transformer for ... | 2022-05-27 | Code |
| 4 | CoCa - Google Brain | 121.69 | No | - | - | - |
| 5 | Microsoft Cognitive Services team | 110.14 | No | VIVO: Visual Vocabulary Pre-Training for Novel O... | 2020-09-28 | - |
| 6 | Single Model | 109.49 | No | SimVLM: Simple Visual Language Model Pretraining... | 2021-08-24 | Code |
| 7 | FudanFVL | 106.55 | No | - | - | - |
| 8 | FudanWYZ | 103.75 | No | - | - | - |
| 9 | Human | 91.62 | No | - | - | - |
| 10 | firethehole | 88.54 | No | - | - | - |
| 11 | IEDA-LAB | 87.51 | No | - | - | - |
| 12 | icgp2ssi1_coco_si_0.02_5_test | 87.15 | No | - | - | - |
| 13 | evertyhing | 85.18 | No | - | - | - |
| 14 | vll@mk514 | 78.91 | No | - | - | - |
| 15 | VinVL (Microsoft Cognitive Services + MSR) | 78.01 | No | VinVL: Revisiting Visual Representations in Visi... | 2021-01-02 | Code |
| 16 | MD | 77.39 | No | - | - | - |
| 17 | RCAL | 75.39 | No | - | - | - |
| 18 | Oscar | 73.75 | No | - | - | - |
| 19 | GRIT (zero-shot, no CBS, no VL pretraining, single model) | 72.6 | No | GRIT: Faster and Better Image captioning Transfo... | 2022-07-20 | Code |
| 20 | ViTCAP-CIDEr-136.7-ENC-DEC-ViTbfocal10-test-CBS | 72.13 | No | - | - | - |
| 21 | vinvl_yuan_cbs | 71.43 | No | - | - | - |
| 22 | UpDown-C | 70.21 | No | - | - | - |
| 23 | Xinyi | 68.92 | No | - | - | - |
| 24 | cxy_nocaps_training | 68.5 | No | - | - | - |
| 25 | UpDown + ELMo + CBS | 66.67 | No | - | - | - |
| 26 | Neural Baby Talk + CBS | 58.48 | No | - | - | - |
| 27 | camel XE | 54.56 | No | - | - | - |
| 28 | ClipCap (MLP + GPT2 tuning) | 49.35 | No | ClipCap: CLIP Prefix for Image Captioning | 2021-11-18 | Code |
| 29 | ClipCap (Transformer) | 49.14 | No | ClipCap: CLIP Prefix for Image Captioning | 2021-11-18 | Code |
| 30 | Neural Baby Talk | 48.73 | No | - | - | - |
| 31 | 7_10-7_40000_predict_test.json | 43.2 | No | - | - | - |
| 32 | Yu-Wu | 39.39 | No | - | - | - |
| 33 | Check | 36.12 | No | - | - | - |
| 34 | nocaps_training | 30.09 | No | - | - | - |
| 35 | UpDown | 30.09 | No | - | - | - |
| 36 | area_attention | 26.55 | No | - | - | - |
| 37 | YX | 26.25 | No | - | - | - |
| 38 | B2 | 25.91 | No | - | - | - |
| 39 | coco_all_19 | 23.07 | No | - | - | - |
| 40 | CS395T | 21.3 | No | - | - | - |