Large scale product recognition challenge

Challenge Overview

Large Scale Product Recognition Challenge

Products-10k production recognition challenge is organized in conjunction with ICPR 2020. Given an image of product, participants are challenged to develop an algorithm recognizing the product at SKU-level. The main difficulties lie in the large amount of products labels, and fine-grained details of similar products. Top-1 and Top-5 accuracy is measured to rank each submission.

Individuals and teams with top submissions will present their work at the workshop.

Competition URL https://www.kaggle.com/c/products-10k

Dataset Overview

Products-10k: Large Scale Product Recognition Dataset

All of the images in our Products-10k dataset are collected from the online shopping website JD.com. Ten thousand frequently bought SKUs are contained in the dataset. All SKUs are organized into a hierarchical structure. There are nearly 190,000 images in total. Respect to the practical application scenes, the distribution of image amount is unbalanced. All images are manually checked/labeled by the production expert team.

Dataset Downloads

Download Products-10K dataset

Products-10k download link: OneDrive or on JD Pan [PassCode: kecp4a](Faster download speed in China) website. The ground-truth labels are also released (including the validation set for Public Leaderboard, and the test set for Private Leaderboard.)

Terms and Conditions

The dataset users have requested permission to use the Products-10k database. In exchange for such permission, the users hereby agree to the following terms and conditions:

The database can only be used for non-commercial research and educational purposes.
The authors of the database make no representations or warranties regarding the Database, including but not limited to warranties of non-infringement or fitness for a particular purpose.
You accepts full responsibility for your use of the Database and shall defend and indemnify the Authors of Products-10k, against any and all claims arising from your use of the Database, including but not limited to your use of any copies of copyrighted images that you may create from the Database.
You may provide research associates and colleagues with access to the Database provided that they first agree to be bound by these terms and conditions.
JD.co. reserve the right to terminate your access to the Database at any time.
If you are employed by a for-profit, commercial entity, your employer shall also be bound by these terms and conditions, and you hereby represents that you are authorized to enter into this agreement on behalf of such employer.
The Intellectual Property Right Law of the People’s Republic of China apply to all disputes under this agreement.