CeRF: Convolutional Neural Radiance Fields for New View Synthesis with Derivatives of Ray Modeling

TL;DR

CeRF introduces a convolutional neural radiance field approach that models derivatives along rays, improving novel view synthesis by reducing geometric ambiguity and enhancing rendering quality.

Contribution

The paper presents a novel convolutional neural radiance field method that models derivatives of radiance along rays, addressing geometric ambiguity and blurring issues in view synthesis.

Findings

Outperforms existing state-of-the-art methods in view synthesis quality

Effectively reduces geometric ambiguity in neural rendering

Demonstrates improved rendering clarity and detail

Abstract

In recent years, novel view synthesis has gained popularity in generating high-fidelity images. While demonstrating superior performance in the task of synthesizing novel views, the majority of these methods are still based on the conventional multi-layer perceptron for scene embedding. Furthermore, light field models suffer from geometric blurring during pixel rendering, while radiance field-based volume rendering methods have multiple solutions for a certain target of density distribution integration. To address these issues, we introduce the Convolutional Neural Radiance Fields to model the derivatives of radiance along rays. Based on 1D convolutional operations, our proposed method effectively extracts potential ray representations through a structured neural network architecture. Besides, with the proposed ray modeling, a proposed recurrent module is employed to solve geometric ambiguity in the fully neural rendering process. Extensive experiments demonstrate the promising results of our proposed model compared with existing state-of-the-art methods.