MonoMIP: Multi-Height Inverse Perspective Mapping for Monocular 3D Object Detection

查看原文 / View Original 编辑 / Edit

Computer Science > Computer Vision and Pattern Recognition arXiv:2610.19071 (cs) [Submitted on 13 Oct 2026] Title:MonoMIP: Multi-Height Inverse Perspective Mapping for Monocular 3D Object Detection Authors:Emir Ulurak, Furkan Kıraç

计算机科学 > 计算机视觉与模式识别 arXiv:2610.19071 (cs) [提交于 2026 年 10 月 13 日] 标题：MonoMIP：用于单目 3D 目标检测的多高度逆透视映射作者：Emir Ulurak, Furkan Kıraç

Abstract: Monocular 3D object detection methods increasingly rely on depth-guided transformers, but they operate entirely in the perspective image plane where spatial relationships are distorted by foreshortening. We propose MonoMIP, a dual-decoder framework that adds complementary 3D geometric reasoning through Bird’s-Eye View (BEV) features constructed via Multi-Height Inverse Perspective Mapping.

摘要：单目 3D 目标检测方法越来越依赖于深度引导的 Transformer，但它们完全在透视图像平面上运行，而空间关系会因透视缩短而扭曲。我们提出了 MonoMIP，这是一个双解码器框架，通过多高度逆透视映射构建的鸟瞰图 (BEV) 特征，增加了互补的 3D 几何推理。

We perform IPM at K elevation planes spanning ground to rooftop, producing an ordered stack of height-conditioned metric BEV features that reduces above-ground misprojection and exposes depth-dependent appearance variation across planes. A BEV decoder with multi-scale deformable attention aggregates this multi-height evidence per query, while a depth-guided 3D decoder provides complementary perspective-space predictions.

我们在从地面到屋顶的 K 个高程平面上执行 IPM，生成一系列高度条件化的度量 BEV 特征堆栈，这减少了地面以上的错误投影，并揭示了跨平面的深度相关外观变化。带有多尺度可变形注意力的 BEV 解码器为每个查询聚合了这种多高度证据，而深度引导的 3D 解码器则提供了互补的透视空间预测。

Without extra training data, MonoMIP attains 25.84% Moderate AP3D|R40 on KITTI val and 21.66% on the official test set for Car, outperforming recent published monocular methods on both the KITTI and Waymo benchmarks.

在没有额外训练数据的情况下，MonoMIP 在 KITTI 验证集上达到了 25.84% 的中等 AP3D|R40，在官方汽车测试集上达到了 21.66%，在 KITTI 和 Waymo 基准测试中均优于近期发表的单目方法。