DQO-MAP: Dual Quadrics Multi-Object mapping with Gaussian Splatting

2025-03-04

AI Summary

To address the low accuracy and poor real-time performance in multi-object pose estimation and reconstruction for robotic object navigation, this paper proposes the first end-to-end multi-object SLAM system integrating dual quadrics parametric modeling with 3D Gaussian splatting. We introduce a novel CPU–GPU collaborative architecture enabling object-ID-driven online extraction, optimization, and semantic association. Crucially, we achieve the first deep coupling of dual quadrics—unifying geometric shape and six-degree-of-freedom pose representation—with high-fidelity Gaussian-splatting-based reconstruction. Evaluated on multiple benchmarks, our method reduces average rotational error by 32%, improves PSNR by 4.1 dB, and increases frame rate by 2.3× over prior approaches. These gains collectively enhance the accuracy, visual quality, and computational efficiency of real-time, dynamic multi-object mapping for robotics.

Technology Category

• Intelligent Robots: State Estimation
• Computer Vision: Multi-modal Vision
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Accurate object perception is essential for robotic applications such as object navigation. In this paper, we propose DQO-MAP, a novel object-SLAM system that seamlessly integrates object pose estimation and reconstruction. We employ 3D Gaussian Splatting for high-fidelity object reconstruction and leverage quadrics for precise object pose estimation. Both of them management is handled on the CPU, while optimization is performed on the GPU, significantly improving system efficiency. By associating objects with unique IDs, our system enables rapid object extraction from the scene. Extensive experimental results on object reconstruction and pose estimation demonstrate that DQO-MAP achieves outstanding performance in terms of precision, reconstruction quality, and computational efficiency. The code and dataset are available at: https://github.com/LiHaoy-ux/DQO-MAP.

Problem

• Accurate object perception for robotic navigation.
• Integration of object pose estimation and reconstruction.
• High-fidelity object reconstruction using Gaussian Splatting.

Innovation

• Integrates object pose estimation and reconstruction
• Uses 3D Gaussian Splatting for object reconstruction
• Leverages quadrics for precise object pose estimation