NeuLF: Efficient Novel View Synthesis with Neural 4D Light Field

TL;DR

NeuLF introduces a neural 4D light field approach for efficient, high-quality novel view synthesis that works with sparse training data and supports applications like auto refocus, outperforming previous methods in complex scenes.

Contribution

The paper proposes a neural implicit 4D light field model that enables efficient and high-quality novel view synthesis with sparse data, outperforming prior dense sampling approaches.

Findings

Achieves state-of-the-art view synthesis quality in complex scenes.

Supports auto refocus through per-ray depth prediction.

Operates at interactive frame rates with small memory footprint.

Abstract

In this paper, we present an efficient and robust deep learning solution for novel view synthesis of complex scenes. In our approach, a 3D scene is represented as a light field, i.e., a set of rays, each of which has a corresponding color when reaching the image plane. For efficient novel view rendering, we adopt a two-plane parameterization of the light field, where each ray is characterized by a 4D parameter. We then formulate the light field as a 4D function that maps 4D coordinates to corresponding color values. We train a deep fully connected network to optimize this implicit function and memorize the 3D scene. Then, the scene-specific model is used to synthesize novel views. Different from previous light field approaches which require dense view sampling to reliably render novel views, our method can render novel views by sampling rays and querying the color for each ray from the network directly, thus enabling high-quality light field rendering with a sparser set of training images. Per-ray depth can be optionally predicted by the network, thus enabling applications such as auto refocus. Our novel view synthesis results are comparable to the state-of-the-arts, and even superior in some challenging scenes with refraction and reflection. We achieve this while maintaining an interactive frame rate and a small memory footprint.