It is well known that the success of machine learning methods on visual recognition tasks is highly dependent on access to large labeled datasets. Unfortunately, performance often drops significantly when the model is presented with data from a new deployment domain which it did not see in training, a problem known as dataset shift. The VisDA challenge aims to test domain adaptation methods’ ability to transfer source knowledge and adapt it to novel target domains.
This year’s challenge focuses on Domain Adaptive Pedestrian Re-identification, where the source and target domains have completely different classes (pedestrian IDs). The particular task is to retrieve the pedestrian instances of the same ID as the query image. This problem is significantly different from previous VisDA challenges, where the source and target domains share some overlapping classes. Moreover, ID matching depends on fine-grained details, making the problem harder than before.
For details and instructions on how to participate, please visit the VisDA challenge website, where you can download the datasets and development kits.
Start: May 1, 2020, midnight
Description: Development phase: create models and submit them or directly submit results on validation data; feed-back are provided on the validation set.
Start: June 26, 2020, midnight
Description: The results on the test set will be revealed when the organizers make them available.
July 25, 2020, 4 p.m.
You must be logged in to participate in competitions.Sign In