pytorch / vision

Datasets, Transforms and Models specific to Computer Vision
https://pytorch.org/vision
BSD 3-Clause "New" or "Revised" License
16.09k stars 6.94k forks source link

CamVid dataset #60

Open souravsingh opened 7 years ago

souravsingh commented 7 years ago

Hi, I am interested in adding CamVid Dataset for Image Segmentation.

The Repository hosting the dataset is here- https://github.com/alexgkendall/SegNet-Tutorial/tree/master/CamVid

cc @pmeier

soumith commented 7 years ago

if you have created a DataLoader for it, please send a Pull Request. Thank you

pmeier commented 3 years ago

Hey @souravsingh, as of torchvision==0.4.0 we support video datasets. Right now we provide UCF101, HMDB51, and Kinetics400.

One could make a case that CamVid is a segmentation task while ours are classification only. While CamVid is not unpopular, it is not as high profile as the built-in ones. We are currently in the process of hashing out criteria for when it is justified to add a new dataset. @fmassa Do we want a video segmentation dataset?