RAVEN: Radar Adaptive Vision Encoders for Efficient Chirp-wise Object Detection and Segmentation

TL;DR

RAVEN is an efficient deep learning architecture for FMCW radar perception that processes raw data in a streaming manner, enabling fast object detection and segmentation with reduced computation.

Contribution

It introduces a novel chirp-wise streaming processing method with an early-exit mechanism, improving efficiency and latency over traditional radar perception models.

Findings

Achieves strong object detection performance on automotive radar benchmarks.

Reduces computation and latency compared to conventional frame-based methods.

Effectively preserves MIMO structure through specialized encoders.

Abstract

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.