Multi-weather vehicle dataset for camouflage generation

This dataset is utilized for adversarial camouflage generation. We collect vehicle datasets in the CARLA simulation environment under 16 weather conditions. These weather conditions are generated by combining four sun altitude angles (-90°, 10°, 45°, 90°) with four fog densities (0, 25, 50, 90). Within each weather scenario,  we randomly choose 16 locations for texture generation. Camera transformation values are randomly selected within specified intervals at each car location. These intervals include four altitude angle intervals ([0, 22.5], [22.5, 45], [45, 67.5], [67.5, 90]), eight azimuth angle intervals([0, 45], [45, 90], ..., [315, 360]), and four distances intervals ([2.5, 7.5], [7.5, 12.5], [12.5, 17.5], [17.5, 22.5]). In total, we employ 32,768 photo-realistic images for texture generation. 

本数据集用于对抗性伪装生成（adversarial camouflage generation）。我们在CARLA仿真环境中采集了16种天气条件下的车辆数据集。这些天气条件由四种太阳高度角（-90°、10°、45°、90°）与四种雾浓度（0、25、50、90）组合生成。在每种天气场景下，我们随机选取16个点位用于纹理生成。在每个车辆点位处，相机的位姿变换参数会在指定区间内随机选取，这些区间包含四种俯仰角区间（[0, 22.5]、[22.5, 45]、[45, 67.5]、[67.5, 90]）、八种方位角区间（[0, 45]、[45, 90]、……、[315, 360]）以及四种距离区间（[2.5, 7.5]、[7.5, 12.5]、[12.5, 17.5]、[17.5, 22.5]）。最终，我们共使用32768张照片级真实感图像用于纹理生成。