Hackathon · DRS hackathon · 2023 · 48 hours · Computer vision
Emotion detector for online classes
A facial-expression model that flags how students are doing during online classes, so teachers know who might need extra support. Built for students with disabilities, in 48 hours.
Who I made it with
AryanTN05, Vinayak Rai, Priyesh Gupta, Rohith Yadav, Aarsh Desai.
What it’s for
The team started from a hard number: according to UNICEF, only 61% of children with disabilities aged 5 to 19 in India attend school. The model watches for emotion and behaviour patterns during online classes and surfaces students who may be struggling.
Why it was hardThe framing mattered as much as the model: the output is a prompt for a teacher to check in, not a grade for a student.
Receipts
- github.com/PlatJack/DRS-Hackathon-2Emotion detector for online classes, aimed at students with disabilities. Kartik is a contributor.
- Previous portfolio, as it stood before this redesign (self-reported)Self-reported: IIIT Dharwad (Data Science & AI), research at IIT Bombay Educational Technology, CRF approach for SSMR.
Next, a little further back in timeHivemind →