NextBestPath: Efficient 3D Mapping of Unseen Environments

TL;DR

This paper introduces NextBestPath, a novel approach for active 3D mapping that predicts long-term goals for efficient scene reconstruction, supported by a new challenging dataset and outperforming existing methods.

Contribution

The paper presents a new long-term goal prediction method for 3D mapping and introduces AiMDoom, a diverse indoor dataset for benchmarking mapping algorithms.

Findings

Outperforms state-of-the-art methods on MP3D and AiMDoom datasets.

Achieves more efficient and comprehensive indoor environment mapping.

Utilizes online data collection, augmentation, and curriculum learning.

Abstract

This work addresses the problem of active 3D mapping, where an agent must find an efficient trajectory to exhaustively reconstruct a new scene. Previous approaches mainly predict the next best view near the agent's location, which is prone to getting stuck in local areas. Additionally, existing indoor datasets are insufficient due to limited geometric complexity and inaccurate ground truth meshes. To overcome these limitations, we introduce a novel dataset AiMDoom with a map generator for the Doom video game, enabling to better benchmark active 3D mapping in diverse indoor environments. Moreover, we propose a new method we call next-best-path (NBP), which predicts long-term goals rather than focusing solely on short-sighted views. The model jointly predicts accumulated surface coverage gains for long-term goals and obstacle maps, allowing it to efficiently plan optimal paths with a unified model. By leveraging online data collection, data augmentation and curriculum learning, NBP significantly outperforms state-of-the-art methods on both the existing MP3D dataset and our AiMDoom dataset, achieving more efficient mapping in indoor environments of varying complexity.