Fast and Flexible Indoor Scene Synthesis via Deep Convolutional Generative Models
Fast and Flexible Indoor Scene Synthesis via Deep Convolutional Generative Models
复制标题
DOI：
10.1109/cvpr.2019.00634
复制
发表时间：
2018-11
期刊：
影响因子：
--
通讯作者：
Daniel Ritchie;Kai Wang;Yu-An Lin
中科院分区：
文献类型：
--
作者：
Daniel Ritchie;Kai Wang;Yu-An Lin
我们为室内场景合成提供了一条新的，快速，灵活的管道，该管道基于深卷积生成模型。我们的方法在基于自上而然的图像表示上运行，并通过通过单独的神经网络模块来预测其类别，位置，方向和大小，将对象迭代插入场景中。我们的管道自然支持部分场景的自动完成以及完整场景的综合，而无需进行任何修改。我们的方法比以前的基于图像的方法要快得多，并且生成的结果优于IT和其他最先进的深层生成场景模型，就训练数据和可感知的视觉质量而言。
We present a new, fast and flexible pipeline for indoor scene synthesis that is based on deep convolutional generative models. Our method operates on a top-down image-based representation, and inserts objects iteratively into the scene by predict their category, location, orientation and size with separate neural network modules. Our pipeline naturally supports automatic completion of partial scenes, as well as synthesis of complete scenes, without any modifications. Our method is significantly faster than the previous image-based method, and generates results that outperforms it and other state-of-the-art deep generative scene models in terms of faithfulness to training data and perceived visual quality.