Multilingual Image-Text Classification 벤치마크
Multilingual Image-Text Classification on GLAMI-1M
Top 1 Accuracy %
- 2022-11-17 — EmbraceNet (image+text): Top 1 Accuracy % 69.7
| Rank | Model | Top 1 Accuracy % | Top 5 Accuracy % | Extra Training Data | Paper | Code | Year |
|---|---|---|---|---|---|---|---|
| 1 | EmbraceNet (image+text) | 69.7 | 94.0 | ✓ | GLAMI-1M: A Multilingual Image-Text Fashion Dataset | glami/glami-1m | 2022 |
| 2 | CLIP (zero-shot image+text) | 32.3 | 74.5 | ✓ | GLAMI-1M: A Multilingual Image-Text Fashion Dataset | glami/glami-1m | 2022 |
| 3 | EmbraceNet (image+text) | 69.7 | 94.0 | ✓ | GLAMI-1M: A Multilingual Image-Text Fashion Dataset | glami/glami-1m | 2022 |
| 4 | CLIP (zero-shot image+text) | 32.3 | 74.5 | ✓ | GLAMI-1M: A Multilingual Image-Text Fashion Dataset | glami/glami-1m | 2022 |
| 5 | EmbraceNet (image+text) | 69.7 | 94.0 | ✓ | GLAMI-1M: A Multilingual Image-Text Fashion Dataset | glami/glami-1m | 2022 |
| 6 | CLIP (zero-shot image+text) | 32.3 | 74.5 | ✓ | GLAMI-1M: A Multilingual Image-Text Fashion Dataset | glami/glami-1m | 2022 |