数据集:
keremberke/valorant-object-detection
任务:
目标检测['dropped spike', 'enemy', 'planted spike', 'teammate']
{'valid': 1983, 'train': 6927, 'test': 988}
pip install datasets
from datasets import load_dataset ds = load_dataset("keremberke/valorant-object-detection", name="full") example = ds['train'][0]
https://universe.roboflow.com/daniels-magonis-0pjzx/valorant-9ufcp/dataset/3
@misc{ valorant-9ufcp_dataset, title = { valorant Dataset }, type = { Open Source Dataset }, author = { Daniels Magonis }, howpublished = { \\url{ https://universe.roboflow.com/daniels-magonis-0pjzx/valorant-9ufcp } }, url = { https://universe.roboflow.com/daniels-magonis-0pjzx/valorant-9ufcp }, journal = { Roboflow Universe }, publisher = { Roboflow }, year = { 2022 }, month = { nov }, note = { visited on 2023-01-27 }, }
CC BY 4.0
This dataset was exported via roboflow.com on December 22, 2022 at 5:10 PM GMT
Roboflow is an end-to-end computer vision platform that helps you
It includes 9898 images. Planted are annotated in COCO format.
The following pre-processing was applied to each image:
No image augmentation techniques were applied.