Charge: A Comprehensive Novel View Synthesis Benchmark and Dataset to Bind Them All

TL;DR

This paper introduces a comprehensive dataset for novel view synthesis derived from a high-quality animated film, featuring diverse modalities and benchmarking scenarios to advance 4D scene reconstruction and view generation models.

Contribution

It provides a new, richly annotated dataset with multiple modalities and benchmarking setups, enabling improved training and evaluation of view synthesis methods.

Findings

Dataset includes high-fidelity RGB, depth, normals, segmentation, and optical flow.

Supports dense, sparse, and monocular benchmarking scenarios.

Facilitates progress in 4D scene reconstruction and view synthesis.

Abstract

This paper presents a new dataset for Novel View Synthesis, generated from a high-quality, animated film with stunning realism and intricate detail. Our dataset captures a variety of dynamic scenes, complete with detailed textures, lighting, and motion, making it ideal for training and evaluating cutting-edge 4D scene reconstruction and novel view generation models. In addition to high-fidelity RGB images, we provide multiple complementary modalities, including depth, surface normals, object segmentation and optical flow, enabling a deeper understanding of scene geometry and motion. The dataset is organised into three distinct benchmarking scenarios: a dense multi-view camera setup, a sparse camera arrangement, and monocular video sequences, enabling a wide range of experimentation and comparison across varying levels of data sparsity. With its combination of visual richness, high-quality annotations, and diverse experimental setups, this dataset offers a unique resource for pushing the boundaries of view synthesis and 3D vision.