Birdsnap is a large bird dataset consisting of 49,829 images from 500 bird species with 47,386 images used for training and 2,443 images used for testing.
Source: Fine-Grained Classification via Mixture of Deep Convolutional Neural Networks
Image Source: http://thomasberg.org/
Variants: Birdsnap
This dataset is used in 2 benchmarks:
| Task | Model | Paper | Date |
|---|---|---|---|
| Image Clustering | TURTLE (CLIP + DINOv2) | Let Go of Your Labels … | 2024-06-11 |
| Fine-Grained Image Classification | NNCLR | With a Little Help from … | 2021-04-29 |
| Fine-Grained Image Classification | EffNet-L2 (SAM) | Sharpness-Aware Minimization for Efficiently Improving … | 2020-10-03 |
| Fine-Grained Image Classification | FixSENet-154 | Fixing the train-test resolution discrepancy | 2019-06-14 |
| Fine-Grained Image Classification | EfficientNet-B7 | EfficientNet: Rethinking Model Scaling for … | 2019-05-28 |
| Fine-Grained Image Classification | GPIPE | GPipe: Efficient Training of Giant … | 2018-11-16 |
Recent papers with results on this dataset: