JavaCV is a Java wrapper library for computer vision and multimedia processing, providing bindings to OpenCV, FFmpeg, and other native libraries via JavaCPP Presets. It offers a unified abstraction layer for video capture (FrameGrabber), recording (FrameRecorder), filtering (FrameFilter), and frame conversion (FrameConverter), with concrete implementations for file, stream, camera, and depth sensor sources. The library also includes advanced computer vision capabilities such as geometric and color calibration, image alignment (Gauss-Newton), projective transformations, blob analysis, marker detection for augmented reality, and OpenCL-accelerated image processing. It is primarily a library (not a standalone application), with sample programs demonstrating real-time video streaming, face detection, object tracking, and projector-camera systems. The codebase is large and well-structured, with abstract base classes and reusable patterns, but shows signs of technical drift due to reliance on deprecated OpenCV APIs (IplImage) and legacy native SDKs. Maintenance is active, with frequent updates to keep pace with upstream library versions.

JavaCPP is a mature, annotation-driven library that provides efficient Java-to-C++ interoperability via JNI. It is not an application but a core infrastructure library used by hundreds of downstream projects (e.g., OpenCV, FFmpeg, TensorFlow bindings) to bridge Java and native C++ code. The system consists of a code generation pipeline (Parser, Generator, Builder) that reads Java annotations and C++ headers to produce JNI wrapper classes, a runtime layer (Loader, Pointer hierarchy) that manages native memory and library loading, and a multidimensional data access layer (Indexer package) optimized for scientific computing. The architecture is modular: build-time code generation (Maven plugin and CLI), runtime library loading with caching and platform detection, and a uniform pointer API with automatic deallocation via PhantomReferences. Key entry points are the Builder class (for build-time code generation) and the Loader class (for runtime library loading). The codebase is largely well-crafted, with reusable patterns like the Indexer factory pattern, PointerScope for deterministic resource management, and annotation-driven configuration. Some core components (Generator, Parser) are acknowledged as complex and messy, reflecting organic growth to handle real-world C++ headers. The repository maintains an extensive test suite and supports a wide range of platforms (Linux, macOS, Windows, Android, iOS, FreeBSD, embedded architectures).