Overview
A hybrid C++ / Java Swing toolkit built on native OpenCV bindings: Haar + DNN face detection, HOG features, contour analysis, camera acquisition with preview and image export.
The problem
Java Swing + JavaCV desktop suite for face detection, feature extraction, and live camera preview pipelines.
My approach
Boring, layered, modular architecture: keep the UI thin, logic pure, storage typed, and the whole thing observable.
Architecture
Tech stack
C++JavaOpenCVJavaCVJNI
Layered nodes
Layered architecture — user-facing UI on top, business logic in the middle, storage and observability on the edges.
Implementation
A hybrid C++ / Java Swing toolkit built on native OpenCV bindings: Haar + DNN face detection, HOG features, contour analysis, camera acquisition with preview and image export.
Results
Live camera preview + face detection + export
What I learned
Write boring code, measure everything, keep the UI thin, and never be afraid to delete features that nobody uses.