To solve the problem that the illumination estimation of 3D scene is easily affected by many factors such as weather conditions and shooting angle,a spherical harmonic coefficient estimation method based on the deep learning was proposed,which included feature extraction module and illumination estimation module.The input image pairs and pre-calculated illumina-tion parameters were used to train the network.The feature extraction module extracted the features of the input image,itera-tively changed the spherical harmonic coefficient loss and rendering loss to optimize the network.The spherical harmonic coe-fficient of the scene illumination was predicted by the illumination estimation module.Compared with other 6 mainstream methods and 3 network models,the evaluation indexes used are significantly reduced.Experimental results show that the pro-posed method can effectively restore the illumination effect,and quantitatively and qualitatively verify that the restored illumina-tion is real in vision.
关键词
深度学习/光照估计/球谐光照/虚实融合/光照一致性/高动态范围/注意力机制
Key words
deep learning/illumination estimation/spherical harmonic illumination/virtual-real fusion/illumination consistency/high dynamic range/attention mechanism
维普
万方数据