Quantifying Facial Age by Posterior of Age Comparisons

Yunxuan Zhang, Li Liu, Cheng Li and Chen-Change Loy

Abstract

We introduce a novel approach for annotating large quantity of in-the-wild facial images with high-quality posterior age distribution as labels. Each posterior provides a probability distribution of estimated ages for a face. Our approach is motivated by observations that it is easier to distinguish who is the older of two people than to determine the person’s actual age. Given a reference database with samples of known ages and a dataset to label, we can transfer reliable annotations from the former to the latter via human-in-the-loop comparisons. We show an effective way to transform such comparisons to posterior via fully-connected and SoftMax layers, so as to permit end-to-end training in a deep network. Thanks to the efficient and effective annotation approach, we collect a new large-scale facial age dataset, dubbed ‘MegaAge’, which consists of 50,000 images. With the dataset, we train a network that jointly performs ordinal hyperplane classification and posterior distribution learning.

Session

Orals - Face Analysis

Files

Paper (PDF)
Supplementary (PDF)

DOI

10.5244/C.31.108
https://dx.doi.org/10.5244/C.31.108

Citation

Bibtex

            @inproceedings{BMVC2017_108,
                title={Quantifying Facial Age by Posterior of Age Comparisons},
                author={Yunxuan Zhang, Li Liu, Cheng Li and Chen-Change Loy},
                year={2017},
                month={September},
                pages={108.1-108.12},
                articleno={108},
                numpages={12},
                booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
                publisher={BMVA Press},
                editor={Tae-Kyun Kim, Stefanos Zafeiriou, Gabriel Brostow and Krystian Mikolajczyk},
                doi={10.5244/C.31.108},
                isbn={1-901725-60-X},
                url={https://dx.doi.org/10.5244/C.31.108}
            }