PACS is an image dataset for domain generalization. It consists of four domains, namely Photo (1,670 images), Art Painting (2,048 images), Cartoon (2,344 images) and Sketch (3,929 images). Each domain contains seven categories.

Source: Deep Domain-Adversarial Image Generation for Domain Generalisation

Homepage

Benchmarks

Add a new result Link an existing benchmark

Task	Dataset Variant	Best Model
Domain Generalization	PACS	SIMPLE+
Annotated Code Search	PACS-StaQC-py	Ensemble:USE-tuned+NCS
Annotated Code Search	PACS-SO-DS	Ensemble:USE-tuned+NCS
Annotated Code Search	PACS-CoNaLa	Ensemble:USE-tuned+NCS
Unsupervised Domain Adaptation	PACS	CoVi
Unsupervised Continual Domain Shift Learning	PACS	CoDAG
Domain Adaptation	PACS	SSGEN