We study the problem of novel view synthesis of objects from a single image. Existing methods have demonstrated the potential in single-view view synthesis. However, they still fail to recover the fine appearance details, especially in self-occluded areas. This is because a single view only provides limited information. We observe that manmade objects usually exhibit symmetric appearances, which introduce additional prior knowledge. Motivated by this, we investigate the potential performance gains of explicitly embedding symmetry into the scene representation. In this paper, we propose SymmNeRF, a neural radiance field (NeRF) based framework that combines local and global conditioning under the introduction of symmetry priors. In particular, SymmNeRF takes the pixel-aligned image features and the corresponding symmetric features as extra inputs to the NeRF, whose parameters are generated by a hypernetwork. As the parameters are conditioned on the image-encoded latent codes, SymmNeRF is thus scene-independent and can generalize to new scenes. Experiments on synthetic and realworld datasets show that SymmNeRF synthesizes novel views with more details regardless of the pose transformation, and demonstrates good generalization when applied to unseen objects. Code is available at: https://github.com/xingyi-li/SymmNeRF.
翻译:我们研究了从单一图像中对天体进行新视角合成的问题。 现有方法在单一视图合成中展示了潜在潜力。 但是,它们仍然未能恢复精细外观细节, 特别是在自我封闭的地区。 这是因为单一视图只提供有限的信息。 我们观察到, 人造天体通常具有对称外观, 并引入更多先前的知识。 我们为此研究将对称性明确嵌入场面表达法的潜在性能收益。 在本文中, 我们提议SymmNERRF, 以神经光亮场( NERF)为基础的框架, 在引入对称前期时结合当地和全球调节。 特别是, SymmNERF 将平准图像特征和相应的对称性特征作为对 NERF 的额外投入, 而后者的参数是由超网络生成的。 由于参数以图像编码潜伏代码为条件, SymmNERF因此是视景为独立的, 可以向新的场景进行概括化。 合成和真实世界数据集的实验显示SymmNERF将S- gregal- developational developations with greal developtions for greabs