I'm looking for a good data set for training a CNN based network to do object localization (i.e. a data set with class labels and bounding box data).

What is a good initial data set to use?

How can I obtain and load all the data (images, labels, bounding boxes)?

I'm using python, so a python-friendly data source is ideal.


The ones that are available out of the box -

  1. Open Images Dataset (Object Detection Track)
  2. MS COCO dataset
  3. Indoor Objects Detection
  4. ILSVRC dataset

Create your own

  1. Search for the object of interest on Google Images (like bikes, cars etc.)
  2. Use the Fatkun Chrome browser plugin to download all images at once
  3. Use LabelImg to annotate and generate a custom dataset
  4. Use pre-trained models like SSD MobileNet to re-train your own models

Might not be an exhaustive one, but it will definitely get you started. Cheers!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.