Knowledge Commons of Institute of Automation,CAS
General vs. Long-Tailed Age Estimation: An Approach to Kill Two Birds With One Stone | |
Bao, Zenghao1,2; Tan, Zichang3![]() ![]() ![]() ![]() ![]() | |
发表期刊 | IEEE TRANSACTIONS ON IMAGE PROCESSING
![]() |
ISSN | 1057-7149 |
2023 | |
卷号 | 32页码:6155-6167 |
摘要 | Facial age estimation has received a lot of attention for its diverse application scenarios. Most existing studies treat each sample equally and aim to reduce the average estimation error for the entire dataset, which can be summarized as General Age Estimation. However, due to the long-tailed distribution prevalent in the dataset, treating all samples equally will inevitably bias the model toward the head classes (usually the adult with a majority of samples). Driven by this, some works suggest that each class should be treated equally to improve performance in tail classes (with a minority of samples), which can be summarized as Long-tailed Age Estimation. However, Long-tailed Age Estimation usually faces a performance trade-off, i.e., achieving improvement in tail classes by sacrificing the head classes. In this paper, our goal is to design a unified framework to perform well on both tasks, killing two birds with one stone. To this end, we propose a simple, effective, and flexible training paradigm named GLAE, which is two-fold. First, we propose Feature Rearrangement (FR) and Pixel-level Auxiliary learning (PA) for better feature utilization to improve the overall age estimation performance. Second, we propose Adaptive Routing (AR) for selecting the appropriate classifier to improve performance in the tail classes while maintaining the head classes. Moreover, we introduce a new metric, named Class-wise Mean Absolute Error (CMAE), to equally evaluate the performance of all classes. Our GLAE provides a surprising improvement on Morph II, reaching the lowest MAE and CMAE of 1.14 and 1.27 years, respectively. Compared to the previous best method, MAE dropped by up to 34%, which is an unprecedented improvement, and for the first time, MAE is close to 1 year old. Extensive experiments on other age benchmark datasets, including CACD, MIVIA, and Chalearn LAP 2015, also indicate that GLAE outperforms the state-of-the-art approaches significantly. |
关键词 | General age estimation long-tailed age estimation class-wise mean absolute error |
DOI | 10.1109/TIP.2023.3327540 |
关键词[WOS] | RECOGNITION |
收录类别 | SCI |
语种 | 英语 |
资助项目 | National Key Research and Development Plan[2021YFE0205700] ; Beijing Natural Science Foundation[JQ23016] ; External Cooperation Key Project of Chinese Academy Sciences[173211KYSB20200002] ; Chinese National Natural Science Foundation Projects[62276254] ; Chinese National Natural Science Foundation Projects[82090051] ; Science and Technology Development Fund of Macau[0070/2020/AMJ] ; Guangdong Provincial Key Research and Development Program[2019B010148001] ; China Computer Federation (CCF)-Zhipu AI Large Model[202219] ; Open Research Projects of Zhejiang Laboratory[2021KH0AB07] ; InnoHK Program |
项目资助者 | National Key Research and Development Plan ; Beijing Natural Science Foundation ; External Cooperation Key Project of Chinese Academy Sciences ; Chinese National Natural Science Foundation Projects ; Science and Technology Development Fund of Macau ; Guangdong Provincial Key Research and Development Program ; China Computer Federation (CCF)-Zhipu AI Large Model ; Open Research Projects of Zhejiang Laboratory ; InnoHK Program |
WOS研究方向 | Computer Science ; Engineering |
WOS类目 | Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic |
WOS记录号 | WOS:001105231300003 |
出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
七大方向——子方向分类 | 模式识别基础 |
国重实验室规划方向分类 | 人工智能基础前沿理论 |
是否有论文关联数据集需要存交 | 否 |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/55223 |
专题 | 多模态人工智能系统全国重点实验室 |
通讯作者 | Wan, Jun |
作者单位 | 1.Chinese Acad Sci CASIA, Inst Automat, State Key Lab Multimodal Artificial Intelligence S, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci UCAS, Sch Artificial Intelligence, Beijing 100049, Peoples R China 3.Baidu Res, Inst Deep Learning, Beijing 100085, Peoples R China 4.Macau Univ Sci & Technol, Sch Comp Sci & Engn, Taipa, Macau, Peoples R China 5.Chinese Acad Sci, Hong Kong Inst Sci & Innovat, Ctr Artificial Intelligence & Robot, Hong Kong, Peoples R China |
第一作者单位 | 中国科学院自动化研究所 |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Bao, Zenghao,Tan, Zichang,Li, Jun,et al. General vs. Long-Tailed Age Estimation: An Approach to Kill Two Birds With One Stone[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2023,32:6155-6167. |
APA | Bao, Zenghao,Tan, Zichang,Li, Jun,Wan, Jun,Ma, Xibo,&Lei, Zhen.(2023).General vs. Long-Tailed Age Estimation: An Approach to Kill Two Birds With One Stone.IEEE TRANSACTIONS ON IMAGE PROCESSING,32,6155-6167. |
MLA | Bao, Zenghao,et al."General vs. Long-Tailed Age Estimation: An Approach to Kill Two Birds With One Stone".IEEE TRANSACTIONS ON IMAGE PROCESSING 32(2023):6155-6167. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
TIP2023-age.pdf(1634KB) | 期刊论文 | 作者接受稿 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
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