S$^3$-MonoDETR：用于单目三维物体检测的监督形状与尺度感知可变形Transformer
S$^3$-MonoDETR: Supervised Shape&Scale-perceptive Deformable Transformer for Monocular 3D Object Detection
AI总结：
本文提出S$^3$-DA模块，通过监督形状与尺度感知的可变形注意力，提升单目三维检测中查询点特征质量，在KITTI和Waymo上实现单次训练下的多类别SOTA性能。
摘要：
近年来，基于Transformer的方法在单目三维物体检测中展现出卓越的性能，能够从单张二维图像预测三维属性。这些方法通常利用视觉和深度表示在物体上生成查询点，其质量对检测精度起着决定性作用。然而，当前Transformer中缺乏几何外观感知的无监督注意力机制，容易为查询点生成噪声特征，严重限制了网络性能，并导致模型在单次训练过程中检测多类别物体的能力较差。为解决这一问题，本文提出了一种新颖的“监督形状与尺度感知可变形注意力”（S$^3$-DA）模块，用于单目三维物体检测。具体而言，S$^3$-DA利用视觉和深度特征生成具有不同形状和尺度的多样化局部特征，并同时预测相应的匹配分布，为每个查询赋予有价值的形状与尺度感知。得益于此，S$^3$-DA能够有效估计属于任何类别的查询点的感受野，使其生成鲁棒的查询特征。此外，我们提出了一种基于多分类的形状与尺度匹配（MSM）损失来监督上述过程。在KITTI和Waymo Open数据集上的大量实验表明，S$^3$-DA显著提高了检测精度，与现有方法相比，在单次训练过程中实现了单类别和多类别三维物体检测的最先进性能。源代码将在https://github.com/mikasa3lili/S3-MonoDETR公开提供。
英文摘要：
Recently, transformer-based methods have shown exceptional performance in monocular 3D object detection, which can predict 3D attributes from a single 2D image. These methods typically use visual and depth representations to generate query points on objects, whose quality plays a decisive role in the detection accuracy. However, current unsupervised attention mechanisms without any geometry appearance awareness in transformers are susceptible to producing noisy features for query points, which severely limits the network performance and also makes the model have a poor ability to detect multi-category objects in a single training process. To tackle this problem, this paper proposes a novel ``Supervised Shape&Scale-perceptive Deformable Attention'' (S$^3$-DA) module for monocular 3D object detection. Concretely, S$^3$-DA utilizes visual and depth features to generate diverse local features with various shapes and scales and predict the corresponding matching distribution simultaneously to impose valuable shape&scale perception for each query. Benefiting from this, S$^3$-DA effectively estimates receptive fields for query points belonging to any category, enabling them to generate robust query features. Besides, we propose a Multi-classification-based Shape&Scale Matching (MSM) loss to supervise the above process. Extensive experiments on KITTI and Waymo Open datasets demonstrate that S$^3$-DA significantly improves the detection accuracy, yielding state-of-the-art performance of single-category and multi-category 3D object detection in a single training process compared to the existing approaches. The source code will be made publicly available at https://github.com/mikasa3lili/S3-MonoDETR.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
