DynIBaR: Neural Dynamic Image-Based Rendering

TL;DR

DynIBaR introduces a volumetric image-based rendering approach that effectively synthesizes high-quality novel views from long, complex dynamic scenes with uncontrolled camera movements, outperforming existing neural radiance field methods.

Contribution

It proposes a scene-motion-aware volumetric rendering framework that improves novel view synthesis for complex, long videos with dynamic scenes and unconstrained camera trajectories.

Findings

Outperforms state-of-the-art dynamic NeRF methods on benchmark datasets.

Produces photo-realistic renderings in challenging real-world scenarios.

Handles complex object motions and camera trajectories effectively.

Abstract

We address the problem of synthesizing novel views from a monocular video depicting a complex dynamic scene. State-of-the-art methods based on temporally varying Neural Radiance Fields (aka dynamic NeRFs) have shown impressive results on this task. However, for long videos with complex object motions and uncontrolled camera trajectories, these methods can produce blurry or inaccurate renderings, hampering their use in real-world applications. Instead of encoding the entire dynamic scene within the weights of MLPs, we present a new approach that addresses these limitations by adopting a volumetric image-based rendering framework that synthesizes new viewpoints by aggregating features from nearby views in a scene-motion-aware manner. Our system retains the advantages of prior methods in its ability to model complex scenes and view-dependent effects, but also enables synthesizing photo-realistic novel views from long videos featuring complex scene dynamics with unconstrained camera trajectories. We demonstrate significant improvements over state-of-the-art methods on dynamic scene datasets, and also apply our approach to in-the-wild videos with challenging camera and object motion, where prior methods fail to produce high-quality renderings. Our project webpage is at dynibar.github.io.