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

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Sen, Anuvab, Mohammad, Mir Sayeed, Mukhopadhyay, Saibal
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914448230318080
author Sen, Anuvab
Mohammad, Mir Sayeed
Mukhopadhyay, Saibal
author_facet Sen, Anuvab
Mohammad, Mir Sayeed
Mukhopadhyay, Saibal
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2604_04490
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle RAVEN: Radar Adaptive Vision Encoders for Efficient Chirp-wise Object Detection and Segmentation
Sen, Anuvab
Mohammad, Mir Sayeed
Mukhopadhyay, Saibal
Signal Processing
Artificial Intelligence
Image and Video Processing
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.
title RAVEN: Radar Adaptive Vision Encoders for Efficient Chirp-wise Object Detection and Segmentation
topic Signal Processing
Artificial Intelligence
Image and Video Processing
url https://arxiv.org/abs/2604.04490