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 4 扩散模型的噪声增强级别选择实验
Tab.4 Selection experiment of noise enhancement level for diffusion model
PB@1B@4MRC
180.538.428.758.3129.5
280.738.728.858.5130.5
380.939.228.858.7132.3
481.239.929.058.9133.8
580.438.228.558.5128.9