How I built a computer vision pipeline that extracts code from skewed textbook photos using 4-point perspective warps, adaptive thresholding, and sandboxed Docker execution.
How I built a real-time biometric attendance system in Python using OpenCV, Dlib’s 68-point landmarks, and 128D ResNet embeddings, solving frame lag and spoofing.
Why jumping to 7B LLMs for search query expansion is overkill, and how our ICICSET 2025 research used unsupervised Latent Dirichlet Allocation to eliminate query drift on low CPU budgets.
NLP-powered search enhancement system for improving query accuracy and relevance
Real-time attendance tracking using face recognition with KNN classification