NOVA: NOvel View Augmentation for Neural Composition of Dynamic Objects

TL;DR

NOVA introduces a new view augmentation method to improve neural radiance fields for realistic 3D dynamic object composition, reducing artifacts and eliminating the need for extra ground truth data.

Contribution

It presents a novel view augmentation strategy that enhances neural radiance fields for dynamic scenes, improving quality and scalability over prior methods.

Findings

Reduces blending artifacts in 3D dynamic object insertion

Achieves comparable PSNR without extra ground truth modalities

Offers a flexible and scalable neural composition framework

Abstract

We propose a novel-view augmentation (NOVA) strategy to train NeRFs for photo-realistic 3D composition of dynamic objects in a static scene. Compared to prior work, our framework significantly reduces blending artifacts when inserting multiple dynamic objects into a 3D scene at novel views and times; achieves comparable PSNR without the need for additional ground truth modalities like optical flow; and overall provides ease, flexibility, and scalability in neural composition. Our codebase is on GitHub.