Multi-view to Novel view: Synthesizing Novel Views with Self-Learned Confidence

We address the task of multi-view novel view synthesis, where we are interested in synthesizing a target image with an arbitrary camera pose from given source images. We propose an end-to-end trainable framework that learns to exploit multiple viewpoints to synthesize a novel view without any 3D supervision. Specifically, our model consists of a flow prediction module and a pixel generation module to directly leverage information presented in source views as well as hallucinate missing pixels from statistical priors. To merge the predictions produced by the two modules given multi-view source images, we introduce a self-learned confidence aggregation mechanism. We evaluate our model on images rendered from 3D object models as well as real and synthesized scenes. We demonstrate that our model is able to achieve state-of-the-art results as well as progressively improve its predictions when more source images are available.

PDF Abstract Proceedings of 2018 PDF Proceedings of 2018 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Novel View Synthesis KITTI Novel View Synthesis Multi-view to Novel View SSIM 0.626 # 1
Novel View Synthesis ShapeNet Car Multi-view to Novel View SSIM 0.923 # 1
Novel View Synthesis ShapeNet Chair Multi-view to Novel View SSIM 0.895 # 1
Novel View Synthesis Synthia Novel View Synthesis Multi-view to Novel View SSIM 0.697 # 1


No methods listed for this paper. Add relevant methods here