UniGeo: A Unified 3D Indoor Object Detection Framework Integrating Geometry-Aware Learning and Dynamic Channel Gating

2026-01-30

AI Summary

Existing methods struggle to effectively model the geometric structure of sparse point clouds and often overlook the feature distribution in critical regions during unified multi-dataset training, thereby limiting the performance of 3D indoor object detection. To address these challenges, this work proposes UniGeo, a novel framework that integrates a geometry-aware learning module with a dynamic channel gating mechanism. The former explicitly enhances feature representations by leveraging spatial geometric relationships, while the latter adaptively refines features produced by a sparse 3D U-Net. Extensive experiments demonstrate that UniGeo significantly outperforms current state-of-the-art approaches across six indoor scene datasets, confirming its strong generalization capability and effectiveness.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Computer Vision: 3D Computer Vision
• Machine Learning: Multimodal Learning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• 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: Sustainability and carbon-aware systems for Web, mobile, and WoT

Abstract

The growing adoption of robotics and augmented reality in real-world applications has driven considerable research interest in 3D object detection based on point clouds. While previous methods address unified training across multiple datasets, they fail to model geometric relationships in sparse point cloud scenes and ignore the feature distribution in significant areas, which ultimately restricts their performance. To deal with this issue, a unified 3D indoor detection framework, called UniGeo, is proposed. To model geometric relations in scenes, we first propose a geometry-aware learning module that establishes a learnable mapping from spatial relationships to feature weights, which enabes explicit geometric feature enhancement. Then, to further enhance point cloud feature representation, we propose a dynamic channel gating mechanism that leverages learnable channel-wise weighting. This mechanism adaptively optimizes features generated by the sparse 3D U-Net network, significantly enhancing key geometric information. Extensive experiments on six different indoor scene datasets clearly validate the superior performance of our method.

Problem

• 3D object detection
• point clouds
• geometric relationships
• feature distribution
• indoor scenes

Innovation

• geometry-aware learning
• dynamic channel gating
• 3D object detection
• point cloud
• unified framework