The Roboflow Model Library contains pre-configured model architectures for easily training computer vision models.
- Description: YOLOv8 is a state-of-the-art object detection and image segmentation model created by Ultralytics.
- Links:
- Description: A very fast and easy to use PyTorch model that achieves state of the art (or near state of the art) results.
- Links:
- Description: The latest in the YOLO mainline, from the creators of YOLOv4, YOLOv7 achieves state of the art performance on MS COCO amongst realtime object detectors.
- Links:
- Description: MT-YOLOv6, or YOLOv6, is a high performance model in the YOLO family of models. Released in June 2022, it sets a new state of the art.
- Links:
- Description: Accelerated Inference of a YOLOv7 Model using OpenVINO™ Integration with Torch-ORT.
- Links:
- Description: As of December 2020, Scaled-YOLOv4 is state-of-the art for object detection. Scaled-YOLOv4 implements YOLOv4 in the PyTorch framework with Cross Stage Partial network layers.
- Links:
- Description: YOLOS is a new transformer based object detection model.
- Links:
- Description: You Only Learn One Representation (YOLOR) is a state-of-the-art object detection model that pre-trains an implicit knowledge network and a set of parameters to represent explicit knowledge.
- Links:
- Description: YOLOX is the winner of the most recent CMU Streaming Perception Challenge for its ability to tradeoff both edge inference speed and accuracy.
- Links:
- Description: A scalable, state of the art object detection model, implemented here within the TensorFlow 2 Object Detection API.
- Links:
- Description: The tiny and fast version of YOLOv4 - good for training and deployment on limited compute resources, and getting a feel for your dataset.
- Links:
- Description: YOLOv4 has emerged as the best real time object detection model. YOLOv4 carries forward many of the research contributions of the YOLO family of models along with new modeling and data augmentation techniques. This implementation is in Darknet.
- Links:
- Description: Oriented bounding boxes are bounding boxes rotated to better fit the objects represented on an angle.
- Links:
- Description: Detectron2 is a model zoo of its own for computer vision models written in PyTorch. Detectron2 includes all the models that were available in the original Detectron, such as Faster R-CNN, Mask R-CNN, RetinaNet, and DensePose. It also features several new models, including Cascade R-CNN, Panoptic FPN, and TensorMask.
- Links:
- Description: EfficientDet achieves the best performance in the fewest training epochs among object detection model architectures, making it a highly scalable architecture especially when operating with limited compute.
- Links:
- Description: YOLOv4 has emerged as one of the best real-time object detection models. YOLOv4 carries forward many of the research contributions of the YOLO family of models along with new modeling and data augmentation techniques. This implementation is in PyTorch.
- Links:
- Description: One of the most accurate object detection algorithms but requires a lot of power at inference time. A good choice if you can do processing asynchronously on a server.
- Links:
- Description: Though it is no longer the most accurate object detection algorithm, YOLO v3 is still a very good choice when you need real-time detection while maintaining excellent accuracy. PyTorch version.
- Links:
- Description: Though it is no longer the most accurate object detection algorithm, YOLO v3 is still a very good choice when you need real-time detection while maintaining excellent accuracy. Keras implementation.
- Links:
- Description: This architecture provides good realtime results on limited compute. It's designed to run in realtime (30 frames per second) even on mobile devices.
- Links: