DATA YOLOv8

该数据集包含可见光与红外热成像双模态视频序列，通过YOLOv8算法实现跨模态特征融合，涵盖复杂光照、遮挡场景下的人体检测任务，提供精准的边界框标注及模态差异标注信息，适用于安防监控、应急救援等领域，可有效提升暗光环境下目标识别鲁棒性。

This dataset contains dual-modal video sequences of visible light and infrared thermal imaging. It leverages the YOLOv8 algorithm to achieve cross-modal feature fusion, covering human detection tasks under complex lighting and occlusion conditions. The dataset provides precise bounding box annotations and modal difference annotation information, which is applicable to fields such as security monitoring and emergency rescue, and can effectively improve the robustness of target recognition in low-light environments.

搜集汇总

背景与挑战

背景概述

DATA YOLOv8数据集包含可见光与红外热成像双模态视频序列，通过YOLOv8算法实现跨模态特征融合，适用于复杂光照、遮挡场景下的人体检测任务，提供精准的边界框标注及模态差异标注信息，可有效提升暗光环境下目标识别鲁棒性。