4K4D: Real-Time 4D View Synthesis at 4K Resolution

TL;DR

This paper introduces 4K4D, a novel 4D point cloud representation enabling real-time, high-quality 4K resolution view synthesis of dynamic scenes by leveraging hardware rasterization and a hybrid appearance model.

Contribution

We propose 4K4D, a 4D point cloud framework with a hybrid appearance model and differentiable depth peeling, achieving unprecedented rendering speed and quality for dynamic scene synthesis.

Findings

Over 400 FPS at 1080p resolution on DNA-Rendering dataset

80 FPS at 4K resolution on ENeRF-Outdoor dataset

30x faster rendering compared to previous methods

Abstract

This paper targets high-fidelity and real-time view synthesis of dynamic 3D scenes at 4K resolution. Recently, some methods on dynamic view synthesis have shown impressive rendering quality. However, their speed is still limited when rendering high-resolution images. To overcome this problem, we propose 4K4D, a 4D point cloud representation that supports hardware rasterization and enables unprecedented rendering speed. Our representation is built on a 4D feature grid so that the points are naturally regularized and can be robustly optimized. In addition, we design a novel hybrid appearance model that significantly boosts the rendering quality while preserving efficiency. Moreover, we develop a differentiable depth peeling algorithm to effectively learn the proposed model from RGB videos. Experiments show that our representation can be rendered at over 400 FPS on the DNA-Rendering dataset at 1080p resolution and 80 FPS on the ENeRF-Outdoor dataset at 4K resolution using an RTX 4090 GPU, which is 30x faster than previous methods and achieves the state-of-the-art rendering quality. Our project page is available at https://zju3dv.github.io/4k4d/.