Animals with Attributes 2
Animals with Attributes 2 (AwA2) is a dataset for benchmarking transfer-learning algorithms, such as attribute base classification and zero-shot learning. AwA2 is a drop-in replacement of original Animals with Attributes (AwA) dataset, with more images released for each category. Specifically, AwA2 consists of in total 37322 images distributed in 50 animal categories. The AwA2 also provides a category-attribute matrix, which contains an 85-dim attribute vector (e.g., color, stripe, furry, size, and habitat) for each category.
Source: Learning from Noisy Web Data with Category-level Supervision
Image Source: https://arxiv.org/pdf/1604.00326.pdf
Variants: AWA2 - 0-Shot, AwA2
This dataset is used in 3 benchmarks:
| Task | Model | Paper | Date |
|---|---|---|---|
| Concept-based Classification | EQ-CBM (ResNet-34) | EQ-CBM: A Probabilistic Concept Bottleneck … | 2024-09-22 |
| Zero-Shot Learning | ZeroDiff | Exploring Data Efficiency in Zero-Shot … | 2024-06-05 |
| Generalized Few-Shot Learning | MVCN | Better Generalized Few-Shot Learning Even … | 2022-11-29 |
| Zero-Shot Learning | DUET (Ours) | DUET: Cross-modal Semantic Grounding for … | 2022-07-04 |
| Zero-Shot Learning | ZSL-KG | Zero-Shot Learning with Common Sense … | 2020-06-18 |
| Generalized Few-Shot Learning | DRAGON | From Generalized zero-shot learning to … | 2020-04-05 |
| Zero-Shot Learning | ZSL_TF-VAEGAN | Latent Embedding Feedback and Discriminative … | 2020-03-17 |
| Generalized Few-Shot Learning | DA-VAE | Generalized Zero- and Few-Shot Learning … | 2019-06-01 |
| Generalized Few-Shot Learning | CA-VAE | Generalized Zero- and Few-Shot Learning … | 2019-06-01 |
| Generalized Few-Shot Learning | CADA-VAE | Generalized Zero- and Few-Shot Learning … | 2018-12-05 |
| Generalized Few-Shot Learning | REVISE | Learning Robust Visual-Semantic Embeddings | 2017-03-17 |
Recent papers with results on this dataset: