Summarize this content to 100 words:
[Submitted on 10 Feb 2026 (v1), last revised 24 Feb 2026 (this version, v3)]
View a PDF of the paper titled Sim2Radar: Toward Bridging the Radar Sim-to-Real Gap with VLM-Guided Scene Reconstruction, by Emily Bejerano and 4 other authors
View PDF
HTML (experimental)
Abstract:Millimeter-wave (mmWave) radar provides reliable perception in visually degraded indoor environments (e.g., smoke, dust, and low light), but learning-based radar perception is bottlenecked by the scarcity and cost of collecting and annotating large-scale radar datasets. We present Sim2Radar, an end-to-end framework that synthesizes training radar data directly from single-view RGB images, enabling scalable data generation without manual scene modeling. Sim2Radar reconstructs a material-aware 3D scene by combining monocular depth estimation, segmentation, and vision-language reasoning to infer object materials, then simulates mmWave propagation with a configurable physics-based ray tracer using Fresnel reflection models parameterized by ITU-R electromagnetic properties. Evaluated on real-world indoor scenes, Sim2Radar improves downstream 3D radar perception via transfer learning: pre-training a radar point-cloud object detection model on synthetic data and fine-tuning on real radar yields up to +3.7 3D AP (IoU 0.3), with gains driven primarily by improved spatial localization. These results suggest that physics-based, vision-driven radar simulation can provide effective geometric priors for radar learning and measurably improve performance under limited real-data supervision.
Submission history From: Emily Bejerano [view email] [v1]
Tue, 10 Feb 2026 10:56:47 UTC (6,682 KB)
[v2]
Fri, 20 Feb 2026 00:41:30 UTC (6,682 KB)
[v3]
Tue, 24 Feb 2026 03:19:01 UTC (6,682 KB)
