Download PDFOpen PDF in browser

Robust Facial Recognition for Occlusions using Facial Landmarks

14 pagesPublished: July 18, 2022

Abstract

Convolutional neural networks have proven to be very powerful for image classification problems, but still has its shortcomings in the presence of non-ideal data. Recently, facial recognition has become popular with usages such as surveillance and automatic tagging of individuals on social media sites. This paper explores a facial recognition solution that utilizes a feature masking strategy focused on facial landmarks with the goal of developing a solution capable of facial recognition in the presence of occlusions. The main driving factor behind this paper is based on the idea that the most commonly found occlusions in the wild are found in the regions of the facial landmarks and that these landmarks play a crucial role during the recognition process. It is found that using a masking strategy based on facial landmarks can be beneficial if the network is trained adequately and the dataset contains mostly well aligned faces, offering improved performance in comparison to using an arbitrary grid layout for all the tested occlusions. Furthermore, it is discovered that masks are not precise at removing the targeted features, causing the masking strategy to also harm recognition process in some cases by accidentally removing critical features.

Keyphrases: CNN, Facial Recognition, Siamese Neural Networks

In: Aurona Gerber (editor). Proceedings of 43rd Conference of the South African Institute of Computer Scientists and Information Technologists, vol 85, pages 48--61

Links:
BibTeX entry
@inproceedings{SAICSIT2022:Robust_Facial_Recognition_for,
  author    = {Kyle Johnston and Mkhuseli Ngxande},
  title     = {Robust Facial Recognition for Occlusions using Facial Landmarks},
  booktitle = {Proceedings of 43rd Conference of the South African Institute of Computer Scientists and Information Technologists},
  editor    = {Aurona Gerber},
  series    = {EPiC Series in Computing},
  volume    = {85},
  pages     = {48--61},
  year      = {2022},
  publisher = {EasyChair},
  bibsource = {EasyChair, https://easychair.org},
  issn      = {2398-7340},
  url       = {https://easychair.org/publications/paper/mMbM},
  doi       = {10.29007/2zjb}}
Download PDFOpen PDF in browser