Description:

The COVID-19 pandemic has necessitated the usage of touch-free solutions in employment settings. Fingerprint sensors are standard practice in the industry and currently require a replacement at short notice. For this purpose, we present an inexpensive, facial recognition login solution using Python.

Modern deep-learning-based facial recognition techniques offer a fast, robust, and accurate solution to this problem and are easy to implement in Python. This talk will focus on the design and implementation of this scalable solution along with presenting a few valuable lessons we learned during the process.

Here is the outline of the talk (Total time - 30 mins)

• Existing Login Practices - 3 mins
• Facial Recognition: past and present - 5 mins
• Solution Design and Implementation - 10 mins
• Testing and Benchmarks - 5 mins
• Lesson Learned - 2 mins
• Q&A - 5 mins

Prerequisites:

Basic understanding of Computer Vision and Deep Learning

Minimal concepts from databases and distributed systems

Video URL:

https://drive.google.com/file/d/1p2kulk9UqeQdNR7ZU4tTOHMDoZmmdh2y/view?usp=sharing

Speaker Info:

Pavan is currently working as a Principal Engineer in OSI Digital, specializing in projects on analytics and cloud applications. He also enjoys contributing to open-source repositories, particularly in machine learning and scientific computing. He has completed his undergraduate degree from IIT Bombay and masters from Stanford University.

Speaker Links:

• Github
• Medium