DetectiumFire
收藏资源简介:
DetectiumFire是一个大规模的多模态数据集,包含了22.5k张高分辨率的火灾相关图像和2.5k个真实世界的火灾相关视频,覆盖了广泛的火灾类型、环境和风险水平。数据集标注了传统的计算机视觉标签(如边界框)和详细的文本提示,描述了场景,支持合成数据生成和火灾风险推理等应用。DetectiumFire在规模、多样性和数据质量方面具有明显优势,显著减少了冗余,并增强了现实场景的覆盖范围。数据集适用于物体检测、基于扩散的图像生成和视觉语言推理等多个任务,有助于推动火灾相关研究和智能安全系统的开发。
DetectiumFire is a large-scale multimodal dataset comprising 22.5k high-resolution fire-related images and 2.5k real-world fire-related videos, covering a wide range of fire types, environments, and risk levels. The dataset is annotated with traditional computer vision labels (e.g., bounding boxes) and detailed textual prompts that describe the scene, supporting applications such as synthetic data generation and fire risk reasoning. DetectiumFire boasts distinct advantages in scale, diversity, and data quality, significantly reducing redundancy and enhancing coverage of real-world scenarios. It supports a variety of tasks including object detection, diffusion-based image generation, and vision-language reasoning, and facilitates the advancement of fire-related research and the development of intelligent safety systems.




