Classify images with two lines of code. These models are ready to go, often with pre-trained weights and exports available for mobile or server-side inference.
If you're more interested in deploying a model without code, check out our Roboflow Deploy product.
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Model Size:
MB
Parameters:
460,000
Top FPS:
Architecture:
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Model Size:
MB
Parameters:
Top FPS:
Architecture:
Model Size:
MB
Parameters:
Top FPS:
Architecture:
CNN
Image classification is a computer vision task where images are assigned a label based on their contents. Only one label is assigned per image. For example, consider a dataset that classifies tree species. One photo may be given the class “birch” and another “fir”.
Image classification is useful in any computer vision task where you need to assign content into one of a limited number of categories. Here are a few examples of real-world use cases for image classification:
There are a wide variety of models used for image classification. Popular choices of models for image classification tasks include YOLOv5, the Vision Transformer, and Resnet34.
Roboflow Universe contains over 100,000 open-source models, many of which you can use for image classification tasks. Below are a few of the many models you can use.
The MIT Indoor Scene Recognition dataset contains 67 categories of indoor scenes on which you can train a classification model. The dataset contains labels ranging from living room to mall to office.
MNIST is a dataset of handwritten digits. The MNIST dataset contains 60,000 training images and 10,000 test images. The digits have been size-normalized and centered in a fixed-size image.
Join 100k developers curating high quality datasets and deploying better models with Roboflow.
Get started
