Research

Applied AI & computer vision

Applied AI and computer vision research focused on face detection and recognition benchmarking — including reproducible evaluation pipelines across pose, age, and resolution.

Publications

Peer-reviewed work in face detection and face recognition benchmarking.

Journal article 2026 · IJCISIM

FaceBench: A Locked-Pipeline Comparative Evaluation of Five Off-the-Shelf Face Recognizers across Pose, Age, and Resolution

Nehalkumar Patel, Bharat Patel

International Journal of Computer Information Systems and Industrial Management Applications · Vol. 18 (19s) · pp. 1221–1235

FaceBench is a reproducible locked-pipeline evaluation of five off-the-shelf face recognizers — FaceNet, Dlib, InsightFace Buffalo-L, AdaFace, and MagFace — under a shared RetinaFace/SCRFD detect-and-crop protocol across pose, age, and resolution benchmarks.

View publication DOI

Conference chapter 2026 · NICE-TEAS ASIA 2025 / Springer

Performance Comparison of Face Detection Algorithms Using the CelebA Dataset

Nehalkumar Patel, Bharat Patel

Expert Systems, Applications and Security (Algorithms for Intelligent Systems), Springer

Comparative evaluation of five face detection algorithms — Dlib-HOG, Haarcascade, SSD, MTCNN, and RetinaFace — on the CelebA dataset using accuracy, precision, recall, F1 score, and false positive rate.

View publication DOI

Focus areas

Computer Vision
Face Detection
Face Recognition
Object Detection
Deep Learning
Video Intelligence

Tools & stacks

Python PyTorch OpenCV YOLO RetinaFace InsightFace FaceNet ArcFace AdaFace MagFace

Research projects

Project case studies and experiment notes will appear here as they are documented.