Datasets
The dog images dataset features 4874 images of dogs in various poses which are encountered during the daily commute in an Indian context. The images have been captured in the IIT Delhi campus using a wearable camera which is used in the MAVI Device. Following are the Annotations details:
Bounding Box
Pose
This dataset contains 1954 images of various signboards placed in the IIT Delhi campus. Annotations contain:
Bounding Box
Text
Direction
Type of signboard
Illumination information
Pedestrain detection for safe mobility is an important concern, hence we have collected the datasets containing pedestrain images from the IIT Delhi campus. This datset cantains 2048 images. Annotations contain
Bounding box
Category
Pose
is_crowd
Visibility
Occlusion
Consists of 2695 images of cows collected across various streets in the city of Delhi. Following annotations are provided:
Bounding Box
Color
Pose
Occluded/Not Occluded
If you wish to use this dataset for research or non-commercial purposes, please feel free to reach out to our group by writing an email to iitd.mavi@gmail.com