基于多视图跨模态特征融合的图像描述生成

张乃洲,赵云超,曹薇,张啸剑

Image captioning generation based on multiple-view cross-modal feature fusion

Naizhou ZHANG,Yunchao ZHAO,Wei CAO,Xiaojian ZHANG

表 1 在MSCOCO测试数据集上与其他先进模型在单一模型上的性能比较

Tab.1 Comparison with other state-of-the-art model on MSCOCO test dataset in single-model setting %

模型 BLEU-1 BLEU-4 METEOR ROUGE-L CIDEr SPICE SCST[6] — 34.2 26.7 55.7 114.0 — AoANet[9] 80.2 38.9 29.2 58.8 129.8 22.4 X-Transformer[10] 80.9 39.7 29.5 59.1 132.8 23.4 M2Transformer[11] 80.8 39.1 29.2 58.6 131.2 22.6 GET[12] 81.5 39.5 29.3 58.9 131.6 22.8 RSTNet[13] 81.8 40.1 29.8 59.5 135.6 23.3 DLCT[14] 81.4 39.8 29.5 59.1 133.8 23.0 Xmodal-Ctx[16] 81.5 39.7 30.0 59.5 135.9 23.7 DIFNet[8] 81.7 40.0 29.7 59.4 136.2 23.2 PureT[21] 82.1 40.9 30.2 60.1 138.2 24.2 VRCDA[33] 80.6 37.9 28.4 58.2 123.7. 21.8 EVCAP[34] — 41.5 31.2 — 140.1 24.7 MVCMFAF (本文) 83.2 41.6 30.4 60.5 140.6 24.4