keremberke/valorant-object-detection

该数据集主要用于目标检测任务，包含四个标签：dropped spike, enemy, planted spike, teammate。数据集包含9898张图像，分为训练集（6927张）、验证集（1983张）和测试集（988张）。图像经过预处理，调整为416x416大小，且未应用图像增强技术。数据集以COCO格式标注，于2022年12月22日通过Roboflow导出。Roboflow是一个端到端的计算机视觉平台，支持团队协作、图像收集与组织、非结构化图像数据理解、标注与数据集创建、模型导出与训练、以及使用主动学习改进数据集。

This dataset is primarily intended for object detection tasks, and includes four labels: dropped spike, enemy, planted spike, and teammate. It contains a total of 9898 images, which are divided into the training set (6927 images), validation set (1983 images) and test set (988 images). All images have been preprocessed and resized to a resolution of 416x416, with no image augmentation techniques applied. The dataset is annotated in COCO format and exported via Roboflow on December 22, 2022. Roboflow is an end-to-end computer vision platform that supports team collaboration, image collection and organization, unstructured image data understanding, annotation and dataset creation, model export and training, as well as dataset improvement using active learning.

原始信息汇总

数据集概述

任务类别

• 目标检测

标签

• dropped spike
• enemy
• planted spike
• teammate

图像数量

• 训练集: 6927张
• 验证集: 1983张
• 测试集: 988张

使用方法

1. 安装 datasets 库: bash pip install datasets

2. 加载数据集: python from datasets import load_dataset ds = load_dataset("keremberke/valorant-object-detection", name="full") example = ds[train][0]

许可证

• CC BY 4.0

数据集处理

• 图像预处理: 调整大小至416x416（拉伸）
• 未应用图像增强技术