Office-Home

ImagesCustom (non-commercial research and educational purposes)Introduced 2017-01-01

Office-Home is a benchmark dataset for domain adaptation which contains 4 domains where each domain consists of 65 categories. The four domains are: Art – artistic images in the form of sketches, paintings, ornamentation, etc.; Clipart – collection of clipart images; Product – images of objects without a background and Real-World – images of objects captured with a regular camera. It contains 15,500 images, with an average of around 70 images per class and a maximum of 99 images in a class.

Source: Multi-component Image Translation for Deep Domain Generalization

Image Source: Wen et al

Benchmarks

Cross-Domain Few-Shot/AUROC Domain Adaptation/Accuracy Domain Adaptation/Avg accuracy Domain Adaptation/Average Accuracy Domain Adaptation/H-Score Domain Adaptation/Source-free Domain Adaptation/VLM Domain Adaptation/Accuracy (%)Domain Generalization/Average Accuracy Few-Shot Learning/AUROC Meta-Learning/AUROC Multi-target Domain Adaptation/Accuracy Transfer Learning/Accuracy Universal Domain Adaptation/H-Score Universal Domain Adaptation/Source-free Universal Domain Adaptation/VLM Unsupervised Domain Adaptation/Accuracy Unsupervised Domain Adaptation/Avg accuracy Unsupervised Domain Adaptation/Average Accuracy

Related Benchmarks

Office-Home (RS-UT imbalance)/Domain Adaptation/Average Per-Class Accuracy Office-Home (RS-UT imbalance)/Unsupervised Domain Adaptation/Average Per-Class Accuracy