Caltech-256 is an object recognition dataset containing 30,607 real-world images, of different sizes, spanning 257 classes (256 object classes and an additional clutter class). Each class is represented by at least 80 images. The dataset is a superset of the Caltech-101 dataset.
Source: Exploiting Non-Linear Redundancy for Neural Model Compression
Image Source: ML4A
Variants: Caltech-256, 1024 Labels, Caltech-256, Caltech-256 5-way (1-shot)
This dataset is used in 2 benchmarks:
| Task | Model | Paper | Date |
|---|---|---|---|
| Image Classification | WaveMixLite-256/7 | WaveMix: A Resource-efficient Neural Network … | 2022-05-28 |
| Image Classification | AG-Net | Attend and Guide (AG-Net): A … | 2021-10-23 |
| Image Classification | Inceptionv4 | Non-binary deep transfer learning for … | 2021-07-19 |
| Image Classification | Inceptionv4 (random initialization) | Non-binary deep transfer learning for … | 2021-07-19 |
| Semi-Supervised Image Classification | UL-Hopfield (ULH) | Unsupervised Learning using Pretrained CNN … | 2018-05-02 |
Recent papers with results on this dataset: