BGG

BGG数据集是由韩国科学技术院（KAIST）的研究团队创建，专注于枪声分类和定位任务。该数据集包含2195个样本，涵盖37种不同类型的枪械、距离和方向信息。数据收集自流行的第一人称射击游戏PUBG，通过模拟真实战场环境来安全地收集枪声数据。BGG数据集不仅有助于游戏中的敌人定位和武器识别，还能提升现实世界中枪声分类和定位任务的准确性，对于军事和警务领域具有重要应用价值。

The BGG dataset was created by a research team from the Korea Advanced Institute of Science and Technology (KAIST), focusing on gunshot classification and localization tasks. It contains 2,195 samples covering 37 distinct types of firearms, as well as distance and direction information. The dataset was collected from the popular first-person shooter game PUBG, where gunshot audio data was safely gathered by simulating realistic battlefield environments. The BGG dataset not only supports enemy localization and weapon recognition in gaming applications, but also improves the accuracy of real-world gunshot classification and localization tasks, possessing significant application value in military and law enforcement domains.