Gentleman_ZH

文章：IEEE 3D Vision 2016的文章 原文ContributionFirst, we introduce a fully convolutional architecture to depth prediction, endowed with novel up-sampling blocks, that allows for dense output maps of hi...

一、将原始数据处理成PyTorch能够加载的数据        imgs为原始的RGB图像，dpts为其对应的深度图。        _path.txt为加载图片的路径，每一行都是：/rgb0.png   /depth0.png class FireWork_Dataset(torch.utils.data.Dataset): def __init__(self, data_...