chainercv
chainercv copied to clipboard
ChainerCV: a Library for Deep Learning in Computer Vision
When I tried [the ImageNet training example](https://github.com/chainer/chainercv/blob/ca45a7ff2f1f6fb45fbc431c8b5bb0f8d646be0f/examples/classification/train_imagenet_multi.py), it failed due to `DirectoryParsingLabelDataset`. ImageNet 2012 dataset has an image with an alpha channel, and `DirectoryParsingLabelDataset` cannot read it. I dealt with...
I tried to train Faster-RCNN from scratch. The train.py gives me this result: { "main/rpn_loc_loss": 0.045347146689891815, "main/rpn_cls_loss": 0.17269396781921387, "main/roi_loc_loss": 0.27309778332710266, "main/roi_cls_loss": 0.49079766869544983, "main/loss": 0.9819365739822388, "lr": 0.0001, "validation/main/map": 0.12205098147362133, "validation/main/ap/aeroplane": 0.2037050348560657,...
based on #880
Issue #139 use package : pykitti package(0.2.4/0.3.0) parseTrackletXML.py (Methods for parsing tracklets (e.g. dataset labels), originally created by Christian Herdtweck.) Function : Display labels and boxes not implemented Grayscale processing
This is a wrapper for the [Visual Genome](http://visualgenome.org/) dataset that I am currently working on. Initially planning to support Region Descriptions (bounding boxes with captions) only but will be grateful...
This PR adds dataset class for OpenImages dataset https://github.com/openimages/dataset This is the first 'working' version and not elaborated at all, so any comments are welcome. :)