SA-Det-100k is a large-scale class-agnostic object detection dataset for Research Purposes only. The dataset is based on a subset of SA-1B (see LICENSE), and all objects belong to the same category objects. Because it contains a large number of scenarios but does not provide class-specific annotations, we believe it may be a good choice to pre-training models for a variety of downstream tasks with different categories. The dataset contains about 100k images, and each image is resized using opencv-python so that the larger one of their height and width is 1333, which is consistent with the data augmentation commonly used to train COCO.
For example project based on this dataset, please see Relation-DETR (https://github.com/xiuqhou/Relation-DETR).
Variants: SA-Det-100k
This dataset is used in 1 benchmark:
Task | Model | Paper | Date |
---|---|---|---|
Object Detection | Relation-DETR (ResNet50 1x) | Relation DETR: Exploring Explicit Position … | 2024-07-16 |
Object Detection | DINO (ResNet50 1x VFL) | DINO: DETR with Improved DeNoising … | 2022-03-07 |
Recent papers with results on this dataset: