Personal tools

Difference between revisions of "Ternary Neural Networks for Face Recognition"

From iis-projects

Jump to: navigation, search
(Literature)
Line 42: Line 42:
 
At the end of the project, you have to present/defend your work during a 15 min. presentation and 5 min. of discussion as part of the IIS colloquium.
 
At the end of the project, you have to present/defend your work during a 15 min. presentation and 5 min. of discussion as part of the IIS colloquium.
  
[[Category:Digital]] [[Category:Available]] [[Category:Semester Thesis]] [[Category:Master Thesis]] [[Category:Bachelor Thesis]] [[Category:Event-Driven Computing]] [[Category:Hot]]
+
[[Category:Digital]] [[Category:Available]] [[Category:Semester Thesis]] [[Category:Master Thesis]] [[Category:Bachelor Thesis]] [[Category:Hot]]
 +
[[Category:Deep Learning Projects]]
 
[[Category:Scheremo]] [[Category:Georg]]
 
[[Category:Scheremo]] [[Category:Georg]]

Revision as of 16:40, 16 July 2021

Introduction

Face recognition is among the many tasks that can be solved with great accuracy using deep learning techniques. Recent work has demonstrated that this task can be successfully tackled with networks using highly quantized (1-bit) weights, known as Binary Weight Networks. This suggests that other aggressively quantized network families may be successfully applied to the problem to increase the energy efficiency of the final system. One such class of networks are ternary neural networks (TNNs), where all weights and activations are quantized to 3 levels: {-1, 0, 1}.

Project description

In this project, you will train a ternary neural network (TNN) for face recognition on the "Labelled Faces in the Wild" (LFW) dataset, optimize it for accuracy and prepare it for deployment on CUTIE, our in-house TNN accelerator.

You will perform the following steps:

  1. Familiarize with the basics of face recognition
  2. Train a baseline full-precision network which can reach good accuracy on face recognition on the LFW dataset
  3. Ternarize and optimize this network
  4. Map the network to CUTIE and perform power simulations to determine the energy cost per inference


Required Skills

  • Basic knowledge of deep learning
  • Basic knowledge of Python

Skills you might find useful, but are not required:

  • Previous experience with highly quantized neural networks

Literature

  • [1] Labelled Faces in the Wild
  • [2] SphereFace: Deep Hypersphere Embedding for Face Recognition
  • [3] P. Jokic et al., Battery-Less Face Recognition at the Extreme Edge
  • [4] M. Scherer et al., CUTIE: Beyond PetaOp/s/W Ternary DNN Inference Acceleration with Better-than-Binary Energy Efficiency

Professor

Luca Benini
Status: Available

Possible to complete as a Semester or Bachelor Thesis

Supervision: Georg Rutishauser georgr@iis.ee.ethz.ch, Moritz Scherer scheremo@iis.ee.ethz.ch

Meetings & Presentations

The students and advisor(s) agree on weekly meetings to discuss all relevant decisions and decide on how to proceed. Of course, additional meetings can be organized to address urgent issues. At the end of the project, you have to present/defend your work during a 15 min. presentation and 5 min. of discussion as part of the IIS colloquium.