Zero-Shot Image Classification

14 benchmarks111 papers

Zero-shot image classification is a technique in computer vision where a model can classify images into categories that were not present during training. This is achieved by leveraging semantic information about the categories, such as textual descriptions or relationships between classes.

Benchmarks

Zero-Shot Image Classification on OVIC Datasets (World-H)

Prediction Score (mean of 3)Overall Score Prediction Score Top 1 Accuracy

Zero-Shot Image Classification on OVIC Datasets (Val3K)

Prediction Score (mean of 3)Top 1 Accuracy (mean of 3)

Zero-Shot Image Classification on OVIC Datasets (Wiki-H)

Overall Score Prediction Score Top 1 Accuracy Prediction Score (mean of 3)

Zero-Shot Image Classification on OVIC Datasets (Wiki-L)

Prediction Score (mean of 3)

Zero-Shot Image Classification on Country211

Zero-Shot Image Classification on ICinW

Zero-Shot Image Classification on ODinW