yankoshbadal/Multi-Face-Attendance-System-using-Deep-Learning
Multi-Face Attendance System using Deep Learning that automatically detects and recognizes multiple individuals simultaneously from images or video streams. The system uses deep learning–based face embeddings to mark attendance accurately and is designed for easy integration with external web or mobile applications.
What's novel
Multi-Face Attendance System using Deep Learning that automatically detects and recognizes multiple individuals simultaneously from images or video streams. The system uses deep learning–based face embeddings to mark attendance accurately and is designed for easy integration with external web or mobile applications.
Code Analysis
14 files read · 3 roundsA CLI-based face attendance system that uses YOLOv8 to detect/crop faces from static images and FaceNet embeddings with cosine similarity to match them against pre-registered student embeddings, writing results to a JSON file.
Strengths
The pipeline is functional and demonstrably works (9 real attendance records in the JSON output), and the docs folder provides unusually thorough line-by-line explanations of each script. The two-pipeline design (build vs. recognize) is a sensible separation of enrollment and recognition concerns.
Weaknesses
Every script hardcodes a Windows-specific absolute path, there are zero tests, and the README claims video stream support, a modular API backend, and AWS/Docker integration that simply do not exist in the code. Significant code duplication between face_detect.py and pre_face_emb.py, no configuration management, and no logging framework further reduce maintainability.
Score Breakdown
Signal breakdown
Innovation
Craft
Traction
Scope
Evidence
Commits
12
Contributors
1
Files
143
Active weeks
2
Repository
Language
Python
Stars
1
Forks
0
License
—