An attendance system built on face recognition is a genuinely useful, well-scoped student project — visual, has real users, and forces engaging with matching thresholds and liveness, not just a demo API call.
The pipeline
- Enrolment: capture several photos per person under varied lighting, not just one ideal shot.
- Matching: embed a live capture and compare against enrolled embeddings with a tuned threshold.
- Liveness: a basic check (blink detection, a prompted head turn) so a printed photo can't fool the system.
- Logging: record the match confidence alongside the attendance entry, so low-confidence matches can be reviewed.
Where student attempts usually cut corners
Skipping liveness detection entirely, and enrolling with a single photo per person rather than several under different conditions — both make the system look fine in a demo and fail in real daily use.
See face-matching thresholds and liveness detection and anti-spoofing basics.
GenAI engineer and AI product builder with 2+ years shipping production-grade AI systems: RAG pipelines, fine-tuned LLMs, hybrid retrieval and multi-modal apps across vision, speech and OCR, architected end to end from ingestion to deployment. Leads engineering for SCULT INDIA's 14-member team, founded the 500+ member TechVerse Enclave community and has mentored 200+ students. Three hackathon first prizes: Changethon 2025 (IIT Roorkee), Product Genesis at Vividhotsava 2025 (CSJMU Kanpur) and BYTEBATTLE (MeetKats).
Open to GenAI roles, hackathon judging, mentorship sessions and guest talks at colleges. Email pranjulrathour41@gmail.com.
