Point2Radio: A Foundation Model for Cross-Scene Radio Fields from Material-Aware Point Clouds

• URL: http://arxiv.org/abs/2607.28994v1
• Date: Fri, 31 Jul 2026 03:45:29 GMT
• Title: Point2Radio: A Foundation Model for Cross-Scene Radio Fields from Material-Aware Point Clouds
• Abstract summary: We present Point2Radio, a foundation model that learns a transferable propagation prior to multiple environments.<n>We evaluate PG prediction on a scene-disjoint split of a 337-scene corpus containing 86,272 TX-conditioned fields.<n>Experiments further show that light target-scene fine-tuning improves adaptation to a specific environment.
• Score: 7.4188535764872485
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: High-fidelity radio fields are typically simulated for every scene--transmitter configuration or fitted separately to each scene, failing to exploit propagation structures shared across environments. We present Point2Radio, a foundation model that learns a transferable propagation prior from multiple environments. Given a material-aware point cloud and a transmitter (TX) setting, a common encoder produces a TX-conditioned scene representation that can be queried at arbitrary receiver (RX) locations. Task-specific query decoders map this representation to different radio quantities, e.g., three-dimensional (3D) path-gain (PG) fields and power angular spectra (PAS). At inference for a new scene, the model uses only a material-aware point cloud and transceiver queries, running in milliseconds on a single GPU without meshes or explicit path tracing. We evaluate PG prediction on a scene-disjoint split of a 337-scene corpus containing 86,272 TX-conditioned fields. Point2Radio achieves 0.871 dB mean absolute error (MAE), reducing error by 76.7% relative to a same-split UNet-style baseline. The same encoder also supports PAS prediction via a task-specific decoder. Experiments further show that light target-scene fine-tuning improves adaptation to a specific environment.

Related papers

• BeamRMX: Radiation-Pattern-Driven Learning for Generalizable Beam Radio Map Prediction and Beam Management [26.316069202754473]
  6G wireless networks are driving larger antenna arrays and highly directional multi-beam transmission.<n>BeamRMX is the first dedicated framework to treat the spatial radiation pattern as the primary BeamRM query.<n>XBase learns multiscale interactions between the radiation query and scene geometry.
  arXiv  Detail & Related papers  (2026-09-01T02:59:08Z)
• GLocFM: A Geometry-Aware Foundation Model for 3D Indoor Wireless Localization [7.4188535764872485]
  We propose GLocFM, a learning-based wireless localizers that exploits WiFi measurements and scene geometry represented as a 3D point cloud.<n>GLocFM reduces mean 3D localization error relative to one of the state-of-the-art localization baselines by 49.5% and 48.8%, respectively.
  arXiv  Detail & Related papers  (2026-08-10T08:39:15Z)
• RadioDiff-v2: Generative Angular Radio Maps for Multi-Beam Selection and Localization [21.594684328904403]
  Angular radio maps describe the received-power distribution over the angle of arrival and underpin beam selection and receiver localization in 6G networks.<n>We propose RadioDiff-v2, a one-dimensional diffusion transformer trained with flow matching.<n>On a zero-shot test of 99 environments and one million links, RadioDiff-v2 leads every baseline on every metric.
  arXiv  Detail & Related papers  (2026-07-09T01:48:47Z)
• U6G XL-MIMO Radiomap Prediction: Multi-Config Dataset and Beam Map Approach [62.11972168232143]
  The upper 6 GHz (U6G) band with XL-MIMO is a key enabler for sixth-generation wireless systems.<n>Existing datasets support only small-scale arrays with predominantly isotropic antennas, far from the 1024-element directional arrays envisioned for 6G.<n>This paper advances XL-MIMO radiomap prediction from three aspects.
  arXiv  Detail & Related papers  (2026-03-06T15:44:27Z)
• Towards Zero-Shot Point Cloud Registration Across Diverse Scales, Scenes, and Sensor Setups [41.033896572767496]
  We present BUFFER-X, a training-free registration framework that achieves zero-shot generalization.<n>Our approach employs hierarchical multi-scale matching to extract correspondences across local, middle, and global receptive fields.<n>For efficiency-critical applications, we introduce BUFFER-X-Lite, which reduces total time by 43%.
  arXiv  Detail & Related papers  (2026-01-06T06:51:24Z)
• RadioDiff-3D: A 3D$\times$3D Radio Map Dataset and Generative Diffusion Based Benchmark for 6G Environment-Aware Communication [76.6171399066216]
  UrbanRadio3D is a large-scale, high-resolution 3D RM dataset constructed via ray tracing in realistic urban environments.<n>RadioDiff-3D is a diffusion-model-based generative framework utilizing the 3D convolutional architecture.<n>This work provides a foundational dataset and benchmark for future research in 3D environment-aware communication.
  arXiv  Detail & Related papers  (2025-07-16T11:54:08Z)
• RadioFormer: A Multiple-Granularity Radio Map Estimation Transformer with 1\textpertenthousand Spatial Sampling [60.267226205350596]
  Radio map estimation aims to generate a dense representation of electromagnetic spectrum quantities.<n>We propose RadioFormer, a novel multiple-granularity transformer to handle the constraints posed by spatial sparse observations.<n>We show that RadioFormer outperforms state-of-the-art methods in radio map estimation while maintaining the lowest computational cost.
  arXiv  Detail & Related papers  (2025-04-27T08:44:41Z)
• RangeSeg: Range-Aware Real Time Segmentation of 3D LiDAR Point Clouds [0.6119392435448721]
  This paper takes advantages of the uneven range distribution of different LiDAR laser beams to propose a range aware instance segmentation network, RangeSeg. Experiments on the KITTI dataset show that RangeSeg outperforms the state-of-the-art semantic segmentation methods with enormous speedup. The whole RangeSeg pipeline meets the real time requirement on NVIDIAtextsuperscripttextregistered JETSON AGX Xavier with 19 frames per second in average.
  arXiv  Detail & Related papers  (2022-05-02T09:57:59Z)
• Adjacent Context Coordination Network for Salient Object Detection in Optical Remote Sensing Images [102.75699068451166]
  We propose a novel Adjacent Context Coordination Network (ACCoNet) to explore the coordination of adjacent features in an encoder-decoder architecture for optical RSI-SOD. The proposed ACCoNet outperforms 22 state-of-the-art methods under nine evaluation metrics, and runs up to 81 fps on a single NVIDIA Titan X GPU.
  arXiv  Detail & Related papers  (2022-03-25T14:14:55Z)
• Neural RF SLAM for unsupervised positioning and mapping with channel state information [51.484516640867525]
  We present a neural network architecture for jointly learning user locations and environment mapping up to isometry. The proposed model learns an interpretable latent, i.e., user location, by just enforcing a physics-based decoder.
  arXiv  Detail & Related papers  (2022-03-15T21:32:44Z)
• Stage-Aware Feature Alignment Network for Real-Time Semantic Segmentation of Street Scenes [59.81228011432776]
  We present a novel Stage-aware Feature Alignment Network (SFANet) for real-time semantic segmentation of street scenes. By taking into account the unique role of each stage in the decoder, a novel stage-aware Feature Enhancement Block (FEB) is designed to enhance spatial details and contextual information of feature maps from the encoder. Experimental results show that the proposed SFANet exhibits a good balance between accuracy and speed for real-time semantic segmentation of street scenes.
  arXiv  Detail & Related papers  (2022-03-08T11:46:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.