Facial Analysis Using Deep Learning

  • Priyanka MoreEmail author
  • Poonam DesaleEmail author
  • Mayuri S. GothwalEmail author
  • Pradnya S. SahajraoEmail author
  • Aarzoo A. ShaikhEmail author
Conference paper
Part of the Lecture Notes on Data Engineering and Communications Technologies book series (LNDECT, volume 33)


A face search system which merges a live search strategy along with a state-of-the-art commercial-off the shelf (COTS) matcher, one cascaded framework. In this first sort massive album of photos, to figure out the top-k most alike faces. The k retrieved prospect is re-ranked by emerging equalities depending on deep features and those results by the COTS matcher. The software based technique is complex and large. It analysis unique shape, pattern and positioning to the respective facial features. It estimates with the records consists of images present in central or local database, the deep network representation combines with a state-of-the-art as well as COTS face matcher in large-scale face search system. According to study on the face datasets leads to complexity: LFW dataset (consist of face detectable). In this project Viola-Jones face detector algorithm is used. The Viola-Jones Technique use to perform feature extraction and evaluation the Rectangular features measures with a new image representation their calculation is very fast.


  1. Chen, B., Chen, Y., Kuo, Y., Hsu, W.: Scalable face image retrieval using attribute-enhanced sparse code-words. IEEE Trans. Multimedia 15(5), 1163–1173 (2012)CrossRefGoogle Scholar
  2. Wu, Z., Ke, Q., Sun, J., Shum, H.Y.: Scalable face image retrieval with identity-based quantization and multi-reference re-rankingGoogle Scholar
  3. Grother, P., Ngan, M.: Face recognition vendor test (FRVT): Performance of face identification algorithms. NIST Interagency, Gaithersburg, Rep. 8009 (2014)Google Scholar
  4. Wang, D., Jain, A.K.: Face retriever: pre-filtering the gallery via deep neural net. In: Proceedings of the International Conference on Biometrics pp. 473–480 (2015)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Genba Sopanrao Moze COEPuneIndia

Personalised recommendations