AliProducts: Large-scale Product Recognition

Challenge Overview

The widespread use of computer vision methods in the retail industry has sparked a new set of challenges, such as the difficult task of correctly distinguishing between a large number of products, especially when the products are quite similar. Tackling this fine-grained classification task may lead to new break-throughs in the field of visual recognition and representation.


The AliProducts dataset consists of ∼3M images of ∼50K different products. The dataset covers many categories of daily commodities, including cosmetics, beverages, snacks, etc. The products are in SKU (Stock Keeping Unit) fine-grained level, and it may be difficult to distinguish between some of the products. Handling class imbalance and noisy training data are also highlights of this challenge.

Challenge Info

This challenge includes a single track, where participants are invited to develop and train their methods using the AliProducts training set and finally test it on the corresponding test set. The challenge winners will receive prizes and may be invited to give a presentation at the CVPR2020 RetailVision workshop.

Prizes: 1500$ for the 1st place, 1000$ for the 2nd place, 500$ for the 3rd place.

For questions about the challenge please contact

Submission and Evaluation

The submission and evaluation procedures are organized and managed in the international competition platform Tianchi. Please see competition page.

Adapted from