ModaNet challenge introduction


#1

If you would like to access to the annotations, please git clone the repo:

https://github.com/eBay/modanet

If you are looking for images, please check the paperdoll repo: https://github.com/kyamagu/paperdoll/tree/master/data/chictopia

Using the command below would help you to get the image_url_list

db = sqlite3.connect('file:./paperdoll/data/chictopia/chictopia.sqlite3?mode=ro', uri=True)
photos = pd.read_sql("""
SELECT
*,
'[http://images2.chictopia.com](http://images2.chictopia.com/)' || path AS url
FROM photos
WHERE photos.post_id IS NOT NULL AND file_file_size IS NOT NULL
""", con=db)

where N from images{N}.[chictopia.com](http://chictopia.com/) can be one of {0, 1, 2, 3}.

For baseline performance, please check our initial technical paper:

Feel free to raise any questions here.


#2

Hi,

I was able to get the annotations. But I’m unable to get the image URLs. When I try to access http://images2.chictopia.com I get an Access denied error.

I’d appreciate pointers on how to get access to the images.

Thanks in advance!


#3

Please refer to the link [https://github.com/kyamagu/paperdoll/tree/master/data/chictopia] pointed in https://github.com/eBay/modanet


#4

Hi,

I am able to get the image url list, but how to match the ‘image_id’ in the annotations?
Thanks.


#5

I think you should be able to get the full image dataset with the image_id. https://github.com/kyamagu/paperdoll/tree/master/data/chictopia