Image Retrieval On Flickr30K 1K Test
评估指标
R@1
R@10
R@5
评测结果
各个模型在此基准测试上的表现结果
比较表格
| 模型名称 | R@1 | R@10 | R@5 |
|---|---|---|---|
| fine-grained-visual-textual-alignment-for | 56.5 | 88.2 | 81.2 |
| dual-attention-networks-for-multimodal | 39.4 | 79.1 | 69.2 |
| visualsparta-sparse-transformer-fragment | 57.4 | 88.1 | 82.0 |
| linking-image-and-text-with-2-way-nets | 36.0 | - | - |
| deep-visual-semantic-alignments-for | 15.2 | 50.5 | - |
| plug-and-play-regulators-for-image-text | 62.6 | 91.1 | 85.8 |
| stacked-cross-attention-for-image-text | 44.0 | 82.6 | 74.2 |
| flickr30k-entities-collecting-region-to | 24.7 | 66.8 | 53.4 |
| learning-semantic-concepts-and-order-for | 41.1 | 80.1 | 70.5 |
| camp-cross-modal-adaptive-message-passing-for | 51.5 | 85.3 | 77.1 |
| multimodal-convolutional-neural-networks-for | 26.2 | 69.6 | 56.3 |
| a-deep-local-and-global-scene-graph-matching | 57.4 | 90.2 | 84.1 |
| learning-deep-structure-preserving-image-text | 29.7 | 72.1 | 60.1 |
| visual-semantic-reasoning-for-image-text | 54.7 | 88.2 | 81.8 |
| instance-aware-image-and-sentence-matching | 30.2 | 72.3 | - |
| multi-grained-vision-language-pre-training | 86.9 | 98.7 | 97.3 |
| similarity-reasoning-and-filtration-for-image | 58.5 | 88.8 | 83.0 |
| fine-grained-visual-textual-alignment-for | 55.7 | 89.3 | 83.1 |