VisDA-2017

Introduced by Peng et al. in VisDA: The Visual Domain Adaptation Challenge

VisDA-2017 is a simulation-to-real dataset for domain adaptation with over 280,000 images across 12 categories in the training, validation and testing domains. The training images are generated from the same object under different circumstances, while the validation images are collected from MSCOCO..

Source: Gradually Vanishing Bridge for Adversarial Domain Adaptation

Homepage

Benchmarks

Add a new result Link an existing benchmark

Task	Dataset Variant	Best Model
Domain Adaptation	VisDA2017	MIC
Universal Domain Adaptation	VisDA2017	LEAD
Source-Free Domain Adaptation	VisDA-2017	SHOT
Unsupervised Domain Adaptation	VisDA2017	SAMB
Partial Domain Adaptation	VisDA2017	SPDA
Semi-supervised Domain Adaptation	VisDA2017	MCL