State of the art object detection models to localize subjects in images. From YOLOv5 to MobileNet, we have the most popular models in easy to use formats.
If you're more interested in deploying a model without code, check out our Roboflow Deploy product.
Model Size:
MB
Parameters:
Top FPS:
Architecture:
YOLO, CNN
Model Size:
14.1
MB
Parameters:
7.2 million
Top FPS:
140
Architecture:
CNN, YOLO
Model Size:
MB
Parameters:
7.5 million
Top FPS:
Architecture:
CNN, YOLO
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Transformers
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Model Size:
68.7
MB
Parameters:
9 million
Top FPS:
Architecture:
CNN, YOLO
Model Size:
75.6
MB
Parameters:
Top FPS:
161
Architecture:
YOLO, CNN
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Model Size:
MB
Parameters:
Top FPS:
Architecture:
YOLO
Model Size:
MB
Parameters:
Top FPS:
520
Architecture:
CNN, YOLO
Model Size:
MB
Parameters:
Top FPS:
Architecture:
YOLO
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Model Size:
202.0
MB
Parameters:
12,786,711 (S2D)
Top FPS:
106
Architecture:
CNN, YOLO
Model Size:
MB
Parameters:
77 million
Top FPS:
8
Architecture:
Model Size:
MB
Parameters:
3.9 million
Top FPS:
97
Architecture:
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Transformer, YOLO
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Model Size:
MB
Parameters:
Top FPS:
34
Architecture:
ResNet-D, YOLO
Object detection is a computer vision solution that focuses on identifying the location of objects in an image or video. Each identified object is assigned a label that represents its contents. Using object detection, you can also count the number of times different objects appear in an image.
Object detection has many use cases, including:
The YOLO family of models (i.e. YOLOv7, YOLOv7) are commonly used in object detection use cases. YOLO has been developed and refined over a years-long period and is still in active development. The latest model, YOLOv7, achieves state-of-the-art performance on object detection in the MS COCO dataset. Other models like Detectron2 and EfficientDet are also used in object detection.
Roboflow Universe contains over 100,000 open-source models, many of which you can use for object detection tasks. Below are a few of the many models you can use.
PlantDoc is a dataset of 2,569 images across 13 plant species and 30 classes (diseased and healthy) for image classification and object detection. There are 8,851 labels.
You can use this model to identify specific diseases in plant materials
This dataset contains 629 annotated images of various scenes on construction sites. The images annotate vehicles, safety hazards (i.e. a worker not wearing a hard hat or vest), and other objects in the environment.
You can use this model to identify safety hazards on a construction site.
Join 100k developers curating high quality datasets and deploying better models with Roboflow.
Get started
