Knowledge Commons of Institute of Automation,CAS
Resizemix: Mixing data with preserved object information and true labels | |
Jie Qin1,2; Jiemin Fang3,4; Qian Zhang5; Wenyu Liu4; Xingang Wang2; Xinggang Wang4 | |
发表期刊 | Computational Visual Media |
2023 | |
页码 | -- |
摘要 | Data augmentation is a powerful technique to increase the diversity of data, which can effectively improve the generalization ability of neural networks in image recognition tasks. Recent mixing-based data augmentations have achieved great success by randomly cropping a patch from one image and pasting it on another. And some works explore to use of the saliency information of the image to guide the mixing. We systematically study the importance of the saliency information for mixing data, and find that the saliency information is not necessary for promoting the augmentation performance. Furthermore, the mixing-based data mixing methods carry two problems of object information missing and label misallocation. We propose an effective and very easily implemented method, namely ResizeMix, which can mix data with preserved object information and true labels. We mix the data by directly resizing the source image to a small patch and paste it on another image. The obtained patch preserves more substantial object information compared with conventional cutting-based methods. ResizeMix achieves superior performance on both image classification and object detection tasks without additional computation cost. |
收录类别 | SCI |
七大方向——子方向分类 | 图像视频处理与分析 |
国重实验室规划方向分类 | 视觉信息处理 |
是否有论文关联数据集需要存交 | 否 |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/57172 |
专题 | 中科院工业视觉智能装备工程实验室_精密感知与控制 |
作者单位 | 1.School of Artificial Intelligence, University of Chinese Academy of Sciences 2.Institute of Automation, Chinese Academy of Sciences 3.Institute of Artificial Intelligence, Huazhong University of Science and Technology 4.School of Electronic Information and Communications, Huazhong University of Science and Technology 5.Horizon Robotics |
第一作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Jie Qin,Jiemin Fang,Qian Zhang,et al. Resizemix: Mixing data with preserved object information and true labels[J]. Computational Visual Media,2023:--. |
APA | Jie Qin,Jiemin Fang,Qian Zhang,Wenyu Liu,Xingang Wang,&Xinggang Wang.(2023).Resizemix: Mixing data with preserved object information and true labels.Computational Visual Media,--. |
MLA | Jie Qin,et al."Resizemix: Mixing data with preserved object information and true labels".Computational Visual Media (2023):--. |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
ResizeMix Mixing Dat(9105KB) | 期刊论文 | 作者接受稿 | 开放获取 | CC BY-NC-SA | 浏览 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论