Institutional Repository of Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
Adaptive Variance Based Label Distribution Learning For Facial Age Estimation | |
Wen Xin1,2; Li Biying1,2; Guo Haiyun1; Liu Zhiwei1; Hu Guosheng3; Tang Ming1; Wang Jinqiao1 | |
2020-08 | |
会议名称 | European Conference on Computer Vision |
会议日期 | 2020-08 |
会议地点 | online |
出版地 | Switzerland |
出版者 | Springer |
摘要 | Estimating age from a single facial image is a classic and
challenging topic in computer vision. One of its most intractable issues
is label ambiguity, i.e., face images from adjacent age of the same person
are often indistinguishable. Some existing methods adopt distribution
learning to tackle this issue by exploiting the semantic correlation be
tween age labels. Actually, most of them set a fifixed value to the variance
of Gaussian label distribution for all the images. However, the variance is
closely related to the correlation between adjacent ages and should vary
across ages and identities. To model a sample-specifific variance, in this pa
per, we propose an adaptive variance based distribution learning (AVDL)
method for facial age estimation. AVDL introduces the data-driven op
timization framework, meta-learning, to achieve this. Specififically, AVDL
performs a meta gradient descent step on the variable (i.e. variance)
to minimize the loss on a clean unbiased validation set. By adaptively
learning proper variance for each sample, our method can approximate
the true age probability distribution more effffectively. Extensive experi
ments on FG-NET and MORPH II datasets show the superiority of our
proposed approach to the existing state-of-the-art methods. |
关键词 | age estimation distribution learning meta-learning |
收录类别 | EI |
语种 | 英语 |
七大方向——子方向分类 | 图像视频处理与分析 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/40581 |
专题 | 模式识别国家重点实验室_图像与视频分析 |
作者单位 | 1.National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences 2.School of Artificial Intelligence, University of Chinese Academy of Sciences 3.Anyvision |
第一作者单位 | 模式识别国家重点实验室 |
推荐引用方式 GB/T 7714 | Wen Xin,Li Biying,Guo Haiyun,et al. Adaptive Variance Based Label Distribution Learning For Facial Age Estimation[C]. Switzerland:Springer,2020. |
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 | ||
adaptive variance ba(2989KB) | 会议论文 | 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