SGDet3D++：面向4D雷达与相机3D目标检测的几何接地语义
SGDet3D++: Geometry-Grounded Semantics for 4D Radar and Camera 3D Object Detection
浏览论文内容
中文总结 AI 辅助
针对4D雷达与相机3D检测中证据支持假设不明确的问题，提出SGDet3D++，通过假设条件证据接地（AGR、GCR、DVC）提升精度，在多个数据集上显著领先。
中文摘要 AI 辅助
4D雷达以长距离几何和径向运动补充了稠密的图像语义，但现有的雷达-相机检测器主要解决模态对齐的“何处”问题，而将“某条证据是否支持一个演化中的目标假设”这一隐含问题搁置。图像标记可能描述遮挡物，附近的雷达回波可能属于另一目标，姿态对齐的记忆槽可能携带不兼容的运动。我们提出了“假设条件证据接地”，将候选访问与证据使用分离：语义、几何或时间证据在更新相应查询之前，由演化中的3D状态进行过滤或条件化。SGDet3D++通过锚点接地语义检索（AGR）实例化这一原则，该机制将可变形图像检索条件化为基于池化的锚点一致雷达支持；几何一致锚点细化（GCR）注意力式地聚合各个关联回波；多普勒验证对应（DVC）仅在当前径向运动与历史矛盾时替换历史。SGDet3D++在OmniHD-Scenes上将最强对比方法提升了3.82 mAP和6.82 ODS，在ManTruckScenes上提升了6.82 mAP和9.22 NDS，同时在TJ4DRadSet测试比较中也领先所列方法。针对机制的评估显示，AGR在每个投影遮挡区间均提升了严格AP，偏航对齐的框门控将目标回波纯度从29.95%提高到58.87%，而DVC保留了96.11%的运动一致历史，同时保持了75.90%的矛盾召回率。代码将发布。
英文摘要
4D radar complements dense image semantics with long-range geometry and radial motion, but existing radar--camera detectors largely solve \emph{where} to align the modalities while leaving \emph{whether} a piece of evidence supports an evolving object hypothesis implicit. An image token may describe an occluder, a nearby radar return may belong to another object, and a pose-aligned memory slot may carry incompatible motion. We formulate \emph{hypothesis-conditioned evidence grounding}, which separates candidate access from evidence use: semantic, geometric, or temporal evidence is filtered or conditioned by the evolving 3D state before updating the corresponding query. \sgdetpp{} instantiates this principle through Anchor-Grounded Semantic Retrieval (AGR), which conditions deformable image retrieval on pooled anchor-consistent radar support; Geometry-Consistent Anchor Refinement (GCR), which attentively aggregates individual associated returns; and Doppler-Verified Correspondence (DVC), which replaces history only when current radial motion contradicts it. \sgdetpp{} improves the strongest compared method by 3.82 mAP and 6.82 ODS on OmniHD-Scenes and by 6.82 mAP and 9.22 NDS on ManTruckScenes, while also leading the listed methods in the TJ4DRadSet test comparison. Mechanism-targeted evaluations show that AGR improves strict AP in every projected-occlusion bin, the yaw-aligned box gate raises target-return purity from 29.95\% to 58.87\%, and DVC preserves 96.11\% of motion-consistent history while retaining 75.90\% contradiction recall. Code will be released.
发表机构
- College of Information Science and Electronic Engineering, Zhejiang University（浙江大学信息科学与电子工程学院）
- School of Automotive Studies, Tongji University（同济大学汽车学院）
机构由 AI 辅助整理，请以论文原文为准。
