Saud Rais

Academic

Career

Verbatim text

SmartEye 
A camera attached on top of a medium-sized ad (bus stop ads, subway ads) sends image frames to 
a server. The server then processes those frames through multiple algorithms to detect and extract 
features from a person’s face. The extracted features are attraction to ad, interest towards ad 
(based on time for which person looks at ad), age group, gender. These features are then used to 
generate a report that helps a company to evaluate the effectiveness of it’s ad placement. 
Features include:  
-Detects and localises head of people passing by the ad 
-Estimates pose of their face to tell where are they looking 
-Computes how many people saw the ad and how many did not 
-Records time for how long each person spent looking at the ad 
-Classifies gender and age of interested persons 
-Based on the above mentioned statistics generates a report which summarizes how successful that 
specific ad is through various visualizations and calculations. 
 
 
 
 
 
 
 
 
 
Technology Used: 
Python, OpenPose, Pytorch, Flask  
ReactJS, Google Colab, VS Code 
Supervisor Name: 
Mr. Shoaib Mehboob 
Group Members:   
Rehan Azhar (i17 - 0003) 
Mohsin Zaidi (i17 - 0216) 
Saud Rais (i17 - 0344)