Proxy embeddings for face identification among multi-pose templates
Weronika Gutfeter , Andrzej Pacut
Many of a large scale face identification systems operates on databases containing images showing heads in multiple poses (from frontal to full profiles). However, as it was shown in the paper, off-the-shelf methods are not able to take advantage of this particular data structure. The main idea behind our work was to adapt the methods proposed for multi-view and semi-3D objects classification to the multi-pose face recognition problem. The proposed approach involves neural network training with proxy embeddings and building the gallery templates out of aggregated samples. A benchmark testing scenario is proposed for the purpose of the problem, which is based on the linked gallery and probes databases. The gallery database consists of multi-pose face images taken under controlled conditions, and the probes database contains samples of in-the-wild type. Both databases must be linked, having at least partially common labels. Two variants of the proposed training procedures were tested, namely, the neighbourhood component analysis with proxies (NCA-proxies) and the triplet margin loss with proxies (triplet-proxies). It is shown that the proposed methods perform better than models trained with cross-entropy loss and than off-the-shelf methods. Rank-1 accuracy was improved from 48.82% for off-the-shelf baseline to 86.86% for NCA-proxies. In addition, transfer of proxy points between two independently trained models was discussed, similarly to hyper-parameters transfer methodology. Proxy embeddings transfer opens a possibility of training two domain-specific networks with respect to two datasets identification schema.
|Journal series||VISIGRAPP 2020 - Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications|
|Score||= 5.0, 28-08-2020, ArticleFromJournal|
|Publication indicators||= 0|
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.