immel-f/KITScenes-Multimodal-Sample-Video

KITScenes多模态样本视频数据集是一个专注于自动驾驶领域的多模态数据集，包含由专家生成的标注，数据规模在100万到1000万之间。它集成了多种传感器数据，如激光雷达、雷达和高清地图（基于Lanelet2格式），适用于机器人学、计算机视觉和多模态人工智能研究。数据集以英语为主要语言，遵循CC BY-NC 4.0许可，并强调隐私保护，通过先进技术匿名化人脸和车牌信息。The KITScenes Multimodal Sample Video dataset is a multimodal dataset focused on autonomous driving scenarios, featuring expert-generated annotations with a size between 1 million and 10 million data points. It integrates multiple sensor data, including LiDAR, radar, and HD maps (in Lanelet2 format), and is designed for robotics, computer vision, and multimodal AI research. The dataset is primarily in English, licensed under CC BY-NC 4.0, and emphasizes privacy protection by anonymizing faces and license plates using state-of-the-art software.