CVC13
收藏资源简介:
CVC-多模态立体数据集由三个多模态立体图像子集 (可见和红外) 组成,每个子集包含不同的场景系列: 道路,立面和光滑表面。每个子集包含在不同条件和日期拍摄的图像。对于每一对多模式立体图像,都会给出一个真实的或合成的深度图,用作地面真实。此外,还提供了相应的校准集。使用具有以下特征的多模态立体声设备记录了图像: 彩色相机: Sony ICX084拜耳模式CCD,6毫米焦距镜头,640 × 480图像分辨率 (大黄蜂2立体声)。红外相机: PathFindIR Flir相机,19毫米焦距镜头 (36度HFOV),534 × 426图像分辨率。所有图像均以无损PPM和BMP格式提供。
The CVC-Multimodal Stereo Dataset consists of three multimodal stereo image subsets (visible and infrared), each containing distinct scene series: roads, facades, and smooth surfaces. Each subset comprises images captured under varying conditions and on different dates. For each pair of multimodal stereo images, a real or synthetic depth map is provided as the ground truth. Additionally, the corresponding calibration set is provided. Images were recorded using a multimodal stereo rig with the following specifications: Color camera: Sony ICX084 Bayer-pattern CCD, 6 mm focal-length lens, 640 × 480 image resolution (Bumblebee 2 Stereo). Infrared camera: PathFindIR Flir camera, 19 mm focal-length lens (36-degree HFOV), 534 × 426 image resolution. All images are provided in lossless PPM and BMP formats.




