- Abstract: We present a novel method for photo-realistic re-rendering of reconstructed objects. The digital reproduction of object appearances is of paramount importance nowadays. Augmented and virtual reality relies on such 3D content. It enables virtual showrooms, virtual tours & sightseeing, the digital inspection of historical artifacts and many other applications. Classical approaches use methods to reconstruct the geometry of an object and textures to capture the appearance properties. Instead, we propose a learned image-guided rendering technique that combines the benefits of image-based rendering and GAN-based image synthesis. A core component of our work is the handling of view-dependent effects. Specifically, we directly train an object-specific deep neural network to synthesize the view-dependent appearance of an object. As input data we are using an RGB video of the object. This video is used to reconstruct a proxy geometry of the object via multi-view stereo. Based on this 3D proxy, the appearance of a captured view can be warped into a new target view. This warping assumes diffuse surfaces, in case of view-dependent effects, such as specular highlights, it leads to artifacts. To this end, we propose EffectsNet, a deep neural network that predicts view-dependent effects. Based on these estimations, we are able to convert observed images to diffuse images. These diffuse images can be projected into other views. In the target view, our pipeline reinserts the new view-dependent effects. To composite multiple reprojected images to a final output, we learn a composition network that outputs photo-realistic results. Using this image-guided approach, the network does not have to allocate capacity on ``''remembering'' object appearance, instead it learns how to combine the appearance of captured images. We demonstrate the effectiveness of our approach both qualitatively and quantitatively on synthetic as well as on real data.
- Keywords: Neural Rendering, Neural Image Synthesis
- TL;DR: We propose a learned image-guided rendering technique that combines the benefits of image-based rendering and GAN-based image synthesis while considering view-dependent effects.