Allen Institute for AI

Segmentation

Image segmentation is the task of segmenting the pixels in an image into a set of object categories. Image segmentation models are typically trained and evaluated on the PASCAL VOC segmentation dataset.

Try it for yourself
1. Upload an Image (or choose one from the examples)
Examples...
Image:
Click to upload your own image
2. Run a model

DeepLabV3

Rethinking Atrous Convolution for Semantic Image SegmentationLiang-Chieh ChenGeorge PapandreouFlorian SchroffHartwig AdamECCV2018

This network employs atrous convolution filters with multiple atrous rates to capture multi-scale context and obtains a mean Intersection over Union (mIOU) of 85.7% on the PASCAL VOC 2012 dataset.