Educational Implementation · JavaScript
fourcast-livive-net
Four visual paradigms, one forecast. An interactive, side-by-side comparison of Computer Vision architectures—HOG+SVM, custom CNN, Transfer Learning (ResNet), and Vision Foundation probes (DINOv2/CLIP)—evaluating cats vs. dogs in real time.
What it is
The repository provides an interactive web application that compares four computer vision architectures (HOG+SVM, custom CNN, transfer learning, and vision foundation models) for cat vs dog classification. It includes code for dataset creation, model training (from-scratch implementations of HOG, SVM, and a custom CNN), model exporting, and a web interface for real-time inference and comparison.
Where it counts
- Computer Vision95%
- Machine Learning85%
- MLOps20%
- Data Science15%
How much of the project is about each field, as scoring counts it. It ranks in the field it is most about.
Around it on the Computer Vision board
Whole board- 1Document-deliveryML Application · Python66.0AdvancedJens MittelbachComputer Vision
- 2AlexNet-from-scratchArchitecture Reimplementation · Jupyter Notebook60.5AdvancedPop AlexandruComputer Vision
- 3fourcast-livive-netEducational Implementation · JavaScript59.1AdvancedJarad PeckenpaughComputer Vision
- 04opencv-document-scannerOther AI Project · Python54.1ImplementationPop AlexandruComputer Vision
- 05FootballNetML Application · Jupyter Notebook48.2ImplementationSuceveanu Mircea Dan SanduComputer Vision
