YOLO TensorFlow ++
A TensorFlow implementation of YOLO for real-time object detection, supporting weight conversion, training, and mobile deployment.
What is YOLO TensorFlow ++?
Darkflow is a TensorFlow port of the YOLO (You Only Look Once) real-time object detection system. It enables users to load pre-trained YOLO weights from the Darknet framework, retrain or fine-tune models with custom datasets, and export optimized graphs for mobile deployment. The project bridges the gap between Darknet's YOLO implementations and the TensorFlow ecosystem.
Machine learning engineers and researchers who want to use YOLO's real-time object detection capabilities within TensorFlow for tasks like custom model training, fine-tuning, and mobile deployment. It's also suitable for developers needing to convert Darknet-trained YOLO models to TensorFlow format.
Darkflow provides a streamlined workflow for transitioning YOLO models from Darknet to TensorFlow, offering unique features like weight conversion, JSON output for detections, and protobuf export for mobile deployment. It simplifies custom training and fine-tuning with configuration files, making YOLO accessible within the TensorFlow toolchain without rewriting models from scratch.
Overview
Translate darknet to tensorflow. Load trained weights, retrain/fine-tune using tensorflow, export constant graph def to mobile devices
Use Cases
Best For
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
