CheapNVS: Real-Time On-Device Narrow-Baseline Novel View Synthesis
This work addresses the challenge of real-time, on-device novel view synthesis for applications like mobile AR/VR, representing a significant incremental improvement in efficiency.
The paper tackles the problem of single-view novel view synthesis by proposing CheapNVS, an efficient method that outperforms state-of-the-art approaches while being 10 times faster and using 6% less memory, achieving real-time performance on mobile devices.
Single-view novel view synthesis (NVS) is a notorious problem due to its ill-posed nature, and often requires large, computationally expensive approaches to produce tangible results. In this paper, we propose CheapNVS: a fully end-to-end approach for narrow baseline single-view NVS based on a novel, efficient multiple encoder/decoder design trained in a multi-stage fashion. CheapNVS first approximates the laborious 3D image warping with lightweight learnable modules that are conditioned on the camera pose embeddings of the target view, and then performs inpainting on the occluded regions in parallel to achieve significant performance gains. Once trained on a subset of Open Images dataset, CheapNVS outperforms the state-of-the-art despite being 10 times faster and consuming 6% less memory. Furthermore, CheapNVS runs comfortably in real-time on mobile devices, reaching over 30 FPS on a Samsung Tab 9+.