eBay eProduct Visual Search Challenge 2022

Overview

Welcome to the 2nd year of eProduct visual search data challenge. Large-scale product recognition is one of the major applications of computer vision and machine learning in the e-commerce domain. Since the number of products is typically much larger than the number of categories of products, image-based product recognition is often cast as a visual search rather than a classification problem. It is also one of the instances of super fine-grained recognition, where there are many products with slight or subtle visual differences.

It has always been a challenge to create a benchmark dataset for training and evaluation on various visual search solutions in a real-world setting. This motivated the creation of eProduct, a large-scale benchmark dataset for product recognition challenges in a visual search setting. The training set contains 1.3M+ listing images, with titles and hierarchical category labels. The evaluation set includes 5K query and 1.1M+ index images for visual search evaluation. You can find additional dataset details and participation information through our eProduct visual search challenge webpage on EvalAI.

The following links share more about the data itself and information about last year’s challenge.

• A white-paper introduces eProduct dataset

• Last year’s final leaderboards and scores.

• Last year’s summary video with brief take-aways.

Competition

Start Date: 1st March 2022

End Date: 1st June 2022

Competition URL: https://eval.ai/web/challenges/challenge-page/1541/overview

Timeline

• 1st March 2022: Beginning of Dev Phase (open for requesting data access)

• 15th May 2022: Dev Phase ends. The last day to update or merge on teams, and to request data access.

• 16th May 2022: Beginning of the Test Phase (release new query data)

• 31st May 2022: Test Phase ends.

• 10th June 2022: Deadline to submit codes and tech report about final solutions.

• 15th June 2022: Final ranking disclosed and winning teams announced.

• 24th June 2022: Workshop day! Winners may present their method depending on review decisions.

Organizers

• Jiangbo Yuan (eBay Inc.)

• An-Ti Chiang (eBay Inc.)

• Wen Tang (eBay Inc.)

• Antonio Haro (eBay Inc.)

Acknowledgements

The eProduct Visual Search Challenge 2022 is sponsored by eBay. We are grateful to many colleagues who provided help launching this challenge. Special thanks to Nitzan Mekel for generous support, Bradley Sanders for legal support, and Tiffany Sandoval for the user agreement process.