Image Paragraph Captioning On Image Paragraph
评估指标
BLEU-4
CIDEr
METEOR
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 | BLEU-4 | CIDEr | METEOR |
|---|---|---|---|
| interactive-key-value-memory-augmented | 10.29 | 24.07 | 17.36 |
| dual-cnn-a-convolutional-language-decoder-for | 8.6 | 17.4 | 15.8 |
| when-an-image-tells-a-story-the-role-of | 4.67 | 26.38 | 11.30 |
| training-for-diversity-in-image-paragraph | 10.58 | 30.63 | 17.86 |
| convolutional-auto-encoding-of-sentence | 9.67 | 25.15 | 18.82 |
| diverse-and-coherent-paragraph-generation | 9.43 | 20.93 | 18.62 |
| hierarchical-scene-graph-encoder-decoder-for | 11.26 | 36.02 | 18.33 |
| recurrent-topic-transition-gan-for-visual | 9.21 | 20.36 | 18.39 |
| look-deeper-see-richer-depth-aware-image | 6.7 | 17.3 | 13.9 |
| a-hierarchical-approach-for-generating | 8.69 | 13.52 | 15.95 |