kang205 / stl-dataset Goto Github PK

Hi is there any way I could get the labels for the bounding boxes... Someone please help me...

Thanks

Hi,
I read your paper and it was very insightful. I was curious about your dataset. However, when I downloaded fashion.json and computed the statistics my numbers are not matching with those of the paper.
I am getting the following numbers:
#scene : 43515 ( In paper it was 47,739 )
#product : 38111
#scene-product pair: 72198

I have used scene_bbox as a primary key for computing the number of unique scenes.
Regards

Hi,

I would like to thank you for sharing your research contributions. Those contributions are very interesting and motivating.
I am trying to conduct research in Image Processing using deep learning for fashion applications (recommendation, STL, and so on); and It would be very useful to have code about the baseline you compared in your article-

Is it possible to get access to your baseline code?

Best,

It looks like you have some issue with assigning a different bounding boxes for the same image.
For example, here's the index image f475bc86db12e719d27a1c440be35690 https://i.pinimg.com/originals/f4/75/bc/f475bc86db12e719d27a1c440be35690.jpg
If you search it in the https://raw.githubusercontent.com/kang205/STL-Dataset/master/fashion.json you'll find that it appears three times with a different corresponding product id but exactly same box [0.555024, 0.857252, 0.734733, 0.952618] which correspond to right lady's shoe.
Same issues could be found for other products (for example 4f2e5549a987f77c260117eed662cd14).
It looks like a result of some map-reduce job where boxes were keyed by product ids instead of product id + scene id