基于跨模态级联扩散模型的图像描述方法

陈巧红,郭孟浩,方贤,孙麒

Image captioning based on cross-modal cascaded diffusion model

Qiaohong CHEN,Menghao GUO,Xian FANG,Qi SUN

表 2 跨模态语义对齐模块在2个数据集上的消融实验

Tab.2 Ablation experiment of cross-modal semantic alignment module in two datasets

ME Linear Microsoft COCO Flickr30k B@1 B@4 M R C B@1 B@4 M R C × × 78.9 34.8 27.5 56.8 116.1 70.1 28.5 22.8 51.2 62.3 √ × 80.5 37.7 28.2 57.3 128.5 72.6 30.1 23.4 51.9 63.8 √ √ 81.2 39.9 29.0 58.9 133.8 74.5 31.2 23.9 53.2 65.4