-
[pdf]
[arXiv]
[bibtex]@InProceedings{Suo_2021_CVPR, author = {Suo, Xin and Jiang, Yuheng and Lin, Pei and Zhang, Yingliang and Wu, Minye and Guo, Kaiwen and Xu, Lan}, title = {NeuralHumanFVV: Real-Time Neural Volumetric Human Performance Rendering Using RGB Cameras}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, month = {June}, year = {2021}, pages = {6226-6237} }
NeuralHumanFVV: Real-Time Neural Volumetric Human Performance Rendering Using RGB Cameras
Abstract
4D reconstruction and rendering of human activities is critical for immersive VR/AR experience. Recent advances still fail to recover fine geometry and texture results with the level of detail present in the input images from sparse multi-view RGB cameras. In this paper, we propose NeuralHumanFVV, a real-time neural human performance capture and rendering system to generate both high-quality geometry and photo-realistic texture of human activities in arbitrary novel views. We propose a neural geometry generation scheme with a hierarchical sampling strategy for real-time implicit geometry inference, as well as a novel neural blending scheme to generate high resolution (e.g., 1k) and photo-realistic texture results in the novel views. Furthermore, we adopt neural normal blending to enhance geometry details and formulate our neural geometry and texture rendering into a multi-task learning framework. Extensive experiments demonstrate the effectiveness of our approach to achieve high-quality geometry and photo-realistic free view-point reconstruction for challenging human performances.
Related Material