ActiveSplat: High-Fidelity Scene Reconstruction through Active Gaussian Splatting

TL;DR

ActiveSplat introduces an autonomous system for high-fidelity scene reconstruction using Gaussian splatting, combining dense and sparse mapping for efficient exploration and photorealistic rendering.

Contribution

It presents a hybrid map representation and hierarchical planning strategy that enhance reconstruction quality and efficiency in autonomous scene mapping.

Findings

Achieves high reconstruction accuracy and completeness.

Demonstrates efficient exploration with promising data coverage.

Validates effectiveness through extensive experiments and ablation studies.

Abstract

We propose ActiveSplat, an autonomous high-fidelity reconstruction system leveraging Gaussian splatting. Taking advantage of efficient and realistic rendering, the system establishes a unified framework for online mapping, viewpoint selection, and path planning. The key to ActiveSplat is a hybrid map representation that integrates both dense information about the environment and a sparse abstraction of the workspace. Therefore, the system leverages sparse topology for efficient viewpoint sampling and path planning, while exploiting view-dependent dense prediction for viewpoint selection, facilitating efficient decision-making with promising accuracy and completeness. A hierarchical planning strategy based on the topological map is adopted to mitigate repetitive trajectories and improve local granularity given limited time budgets, ensuring high-fidelity reconstruction with photorealistic view synthesis. Extensive experiments and ablation studies validate the efficacy of the proposed method in terms of reconstruction accuracy, data coverage, and exploration efficiency. The released code will be available on our project page: https://li-yuetao.github.io/ActiveSplat/.