NCVPRIPG 2019 IDD Benchmark

Semantic segmentation on IDD Lite Dataset

This challenge is open to students in Indian universities and colleges. The challenge will use IDD-Lite, a dataset that is small and compact to fit on any personal computer and so will not huge-compute infrastructure. IDD-Lite is less than 50MB is size, contains 7 classes (compared to 30 in IDD).

Directions for Participation

1. Register here for receiving updates about the competition.
2. Register an account at the IDD website: http://idd.insaan.iiit.ac.in/
3. After logging in to IDD website, download the IDD Lite dataset (http://idd.insaan.iiit.ac.in/dataset/download/)
4. Segmentations masks have all been already generated for IDD Lite in the download.
5. Once you have built a model, and have the predictions of the model in any of the split (train, val), you can evaluate the metric as directed here: https://github.com/AutoNUE/public-code#evaluation. Use the following command for segmentation evaluation:
   python evaluate/idd_lite_evaluate_mIoU.py --gts $GT --preds $PRED --num-workers $C
   Your predictions is a png image, which has the size of 256x128. Each pixel of this image contains the label as level1Ids (see labels code) of the corresponding image (resized to 256x128). The evaluation code above resizes both your prediction and ground truth png files to 256x128, in case they are not of that size.
6. Finally you can upload the predictions for the test split, to be evaluated for the leaderboard here: http://idd.insaan.iiit.ac.in/evaluation/submission/submit/

Results