Compact and adaptive multiplane images for view synthesis

TL;DR

This paper introduces a learning-based method to generate compact, adaptive multiplane images that optimize memory usage and scene geometry for efficient view synthesis.

Contribution

It presents a novel approach to produce memory-efficient MPIs that adaptively sample depths based on scene geometry, improving practicality for applications.

Findings

Reduced memory requirements for MPIs

Maintained high-quality view synthesis

Adaptive depth sampling improves scene representation

Abstract

Recently, learning methods have been designed to create Multiplane Images (MPIs) for view synthesis. While MPIs are extremely powerful and facilitate high quality renderings, a great amount of memory is required, making them impractical for many applications. In this paper, we propose a learning method that optimizes the available memory to render compact and adaptive MPIs. Our MPIs avoid redundant information and take into account the scene geometry to determine the depth sampling.