The Places365 dataset is a scene recognition dataset. It is composed of 10 million images comprising 434 scene classes. There are two versions of the dataset: Places365-Standard with 1.8 million train and 36000 validation images from K=365 scene classes, and Places365-Challenge-2016, in which the size of the training set is increased up to 6.2 million extra images, including 69 new scene classes (leading to a total of 8 million train images from 434 scene classes).
Source: Semantic-Aware Scene Recognition
Image Source: Places
Variants: Places365-Standard, Places365, ImageNet-1k vs Places, ImageNet-1k vs Curated OODs (avg.)
This dataset is used in 2 benchmarks:
| Task | Model | Paper | Date |
|---|---|---|---|
| Image Classification | OmniVec(ViT) | OmniVec: Learning robust representations with … | 2023-11-07 |
| Image Classification | ViC-MAE (ViT-L) | ViC-MAE: Self-Supervised Representation Learning from … | 2023-03-21 |
| Image Classification | InternImage-H（CNN） | InternImage: Exploring Large-Scale Vision Foundation … | 2022-11-10 |
| Image Classification | µ2Net+ (ViT-L/16) | A Continual Development Methodology for … | 2022-09-15 |
| Image Classification | MixMIM-L(ViT-L) | MixMAE: Mixed and Masked Autoencoder … | 2022-05-26 |
| Image Classification | MixMIM-B (ViT) | MixMAE: Mixed and Masked Autoencoder … | 2022-05-26 |
Recent papers with results on this dataset: