Few-shot learning with saliency maps as additional visual information.评价结果

评估详情