UniDet3D：多数据集室内三维物体检测

UniDet3D: Multi-dataset Indoor 3D Object Detection

September 6, 2024

作者: Maksim Kolodiazhnyi, Anna Vorontsova, Matvey Skripkin, Danila Rukhovich, Anton Konushin

cs.AI

摘要

随着客户对机器人技术和增强现实智能解决方案的需求不断增长，对来自点云的三维物体检测引起了相当大的关注。然而，现有的室内数据集单独采集的数据量太小，样本多样性不足，无法训练出强大且通用的三维物体检测模型。与此同时，更通用的方法利用基础模型仍然质量不如基于特定任务的监督训练。在这项工作中，我们提出了一种简单而有效的三维物体检测模型，该模型在多个室内数据集的混合数据上进行训练，能够在各种室内环境中运行。通过统一不同的标签空间，使得模型能够通过监督联合训练方案在多个数据集上学习强大的表示。所提出的网络架构建立在一个基本的Transformer编码器之上，使得对于实际应用来说，运行、定制和扩展预测流程变得更加容易。大量实验证明，在6个室内基准测试中，相较于现有的三维物体检测方法，该模型取得了显著的进展：ScanNet（+1.1 mAP50）、ARKitScenes（+19.4 mAP25）、S3DIS（+9.1 mAP50）、MultiScan（+9.3 mAP50）、3RScan（+3.2 mAP50）和ScanNet++（+2.7 mAP50）。代码可在 https://github.com/filapro/unidet3d 获取。

English

Growing customer demand for smart solutions in robotics and augmented reality has attracted considerable attention to 3D object detection from point clouds. Yet, existing indoor datasets taken individually are too small and insufficiently diverse to train a powerful and general 3D object detection model. In the meantime, more general approaches utilizing foundation models are still inferior in quality to those based on supervised training for a specific task. In this work, we propose , a simple yet effective 3D object detection model, which is trained on a mixture of indoor datasets and is capable of working in various indoor environments. By unifying different label spaces, enables learning a strong representation across multiple datasets through a supervised joint training scheme. The proposed network architecture is built upon a vanilla transformer encoder, making it easy to run, customize and extend the prediction pipeline for practical use. Extensive experiments demonstrate that obtains significant gains over existing 3D object detection methods in 6 indoor benchmarks: ScanNet (+1.1 mAP50), ARKitScenes (+19.4 mAP25), S3DIS (+9.1 mAP50), MultiScan (+9.3 mAP50), 3RScan (+3.2 mAP50), and ScanNet++ (+2.7 mAP50). Code is available at https://github.com/filapro/unidet3d .