Skip to content
EduSense: Practical Classroom Sensing at Scale
Python Go Dockerfile Shell
Branch: master
Clone or download
Type Name Latest commit message Commit time
Failed to load latest commit information.
compose add area of interest box Nov 21, 2019
compute add area of interest box Nov 21, 2019
doc Fix broken links in /doc/ Nov 19, 2019
media added media Nov 11, 2019
storage update copyright/license info Nov 16, 2019
.gitignore Initial commit Oct 10, 2019
.gitmodules Initial commit Oct 10, 2019
LICENSE Adding the BSD 3-clause license Nov 6, 2019 update copyright/license info Nov 16, 2019

EduSense: Practical Classroom Sensing at Scale

hero image

EduSense represents the first real-time, in-the-wild evaluated and practically-deployable classroom sensing system at scale that produces a plethora of theoretically-motivated visual and audio features correlated with effective instruction.

Our getting started is a good starting point if you are interested in building/developing/deploying EduSense. More information about the team can be found on the EduSense website.


  • Oct 2019 We open-source our EduSense code!
  • Sep 2019 We presented our paper titled "Edusense: Practical Classroom Sensing at Scale" at Ubicomp'19.

Features for Students and Instructors


  • Visual Features:
    • Body Segmentation, Keypoints and Inter-frame tracking:
      • Hand Raise Detection
      • Upper Body Pose Estimation
      • Sit vs Stand Detection
      • Synthetic Accelerometer
      • Classroom Topology
    • Facial Lanndmarks and Attributes:
      • Smile Detection
      • Mouth State Detection
      • Gaze Estimation
  • Audio Features:
    • Speech Detection:
      • Student vs Instructor Speech
      • Speech Act Delimation

Visualization Dashboard

viz dashboard

System Architecture

system architecture

Related Links


Karan Ahuja, Dohyun Kim, Franceska Xhakaj, Virag Varga, Anne Xie, Stanley Zhang, Jay Eric Townsend, Chris Harrison, Amy Ogan, and Yuvraj Agarwal. 2019. EduSense: Practical Classroom Sensing at Scale. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. 3, 3, Article 71 (September 2019), 26 pages. DOI:

 author = {Ahuja, Karan and Kim, Dohyun and Xhakaj, Franceska and Varga, Virag and Xie, Anne and Zhang, Stanley and Townsend, Jay Eric and Harrison, Chris and Ogan, Amy and Agarwal, Yuvraj},
 title = {EduSense: Practical Classroom Sensing at Scale},
 journal = {Proc. ACM Interact. Mob. Wearable Ubiquitous Technol.},
 issue_date = {September 2019},
 volume = {3},
 number = {3},
 month = sep,
 year = {2019},
 issn = {2474-9567},
 pages = {71:1--71:26},
 articleno = {71},
 numpages = {26},
 url = {},
 doi = {10.1145/3351229},
 acmid = {3351229},
 publisher = {ACM},
 address = {New York, NY, USA},
 keywords = {Audio, Classroom, Computer Vision, Instructor, Machine Learning, Pedagogy, Sensing, Speech Detection, Teacher},


The source code in this directory and its subdirectories are all governed by BSD 3-Clause License unless otherwise noted in the source code. Once compiled or packaged, it is the user's reponsibility to ensure that any use of the result binary or image complies with any relevant licenses for all software packaged together.

You can’t perform that action at this time.