RAVEN通过分片处理雷达数据，实现低延迟高精度目标检测与分割。
RAVEN: Radar Adaptive Vision Encoders for Efficient Chirp-wise Object Detection and Segmentation
- 按回波片段流式处理原始雷达数据，保留多天线结构
- 引入可学习的跨天线混合模块，压缩虚拟阵列特征
- 支持早期退出，仅用部分回波即可决策，适合车载实时场景
本文提出RAVEN，一种面向调频连续波（FMCW）雷达感知的高效深度学习架构。该方法以分片流式方式处理原始ADC数据，通过独立接收机状态空间编码器保持MIMO结构，并引入可学习的跨天线混合模块，恢复紧凑的虚拟阵列特征。同时设计了早期退出机制，当潜在状态稳定时，模型可仅基于部分回波做出判断。在多个车载雷达基准测试中，该方法在目标检测与鸟瞰图（BEV）空域分割任务上表现优异，且显著降低计算量与端到端延迟，优于传统帧式雷达处理流程。
原文摘要 · Abstract (English)
This paper presents RAVEN, a computationally efficient deep learning architecture for FMCW radar perception. The method processes raw ADC data in a chirp-wise streaming manner, preserves MIMO structure through independent receiver state-space encoders, and uses a learnable cross-antenna mixing module to recover compact virtual-array features. It also introduces an early-exit mechanism so the model can make decisions using only a subset of chirps when the latent state has stabilized. Across automotive radar benchmarks, the approach reports strong object detection and BEV free-space segmentation performance while substantially reducing computation and end-to-end latency compared with conventional frame-based radar pipelines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。