• Corpus ID: 232417078

Multi-View Radar Semantic Segmentation

  title={Multi-View Radar Semantic Segmentation},
  author={Arthur Ouaknine and Alasdair Newson and Patrick P'erez and Florence Tupin and Julien Rebut},
Understanding the scene around the ego-vehicle is key to assisted and autonomous driving. Nowadays, this is mostly conducted using cameras and laser scanners, despite their reduced performance in adverse weather conditions. Automotive radars are low-cost active sensors that measure properties of surrounding objects, including their relative speed, and have the key advantage of not being impacted by rain, snow or fog. However, they are seldom used for scene understanding due to the size and… 
2 Citations
Deep Instance Segmentation with High-Resolution Automotive Radar
Two efficient methods for instance segmentation with radar detection points are proposed, one is implemented in an end-to-end deep learning driven fashion using PointNet++ framework, and the other is based on clustering of the Radar detection points with semantic information.
Raw High-Definition Radar for Multi-Task Learning
A novel HD radar sensing model, FFT-RadNet, is proposed, that eliminates the overhead of computing the Range-Azimuth-Doppler 3D tensor, learning instead to recover angles from a Range- doppler spectrum.


CARRADA Dataset: Camera and Automotive Radar with Range- Angle- Doppler Annotations
This work introduces CARRADA, a dataset of synchronized camera and radar recordings with range-angle-Doppler annotations, and presents a semi-automatic annotation approach, which was used to annotate the dataset, and a radar semantic segmentation baseline, which is evaluated on several metrics.
RSS-Net: Weakly-Supervised Multi-Class Semantic Segmentation with FMCW Radar
This paper presents an efficient annotation procedure and an application thereof to end-to-end, rich semantic segmentation of the sensed environment using Frequency-Modulated Continuous-Wave scanning
RADIATE: A Radar Dataset for Automotive Perception
This paper presents the RAdar Dataset In Adverse weather (RADIATE), aiming to facilitate research on object detection, tracking and scene understanding using radar sensing for safe autonomous driving, and is the first public radar dataset which provides high-resolution radar images on public roads with a large amount of road actors labelled.
Probabilistic Oriented Object Detection in Automotive Radar
This paper proposes a deeplearning based algorithm that takes in radar data in its raw tensor representation and places probabilistic oriented bounding boxes around the detected objects in bird’s-eye-view space and develops a vehicle detection pipeline using raw radar data as the only input.
Road Scene Understanding by Occupancy Grid Learning from Sparse Radar Clusters using Semantic Segmentation
This work proposes learning the inverse sensor model used for occupancy grid mapping from clustered radar data, and shows both qualitatively and quantitatively that the learned occupancy net outperforms classic methods by a large margin using the recently released NuScenes real-world driving data.
Scene Understanding With Automotive Radar
This work presents a complete pipeline to obtain semantic information for each target measured by a network of radar sensors and develops a new set of layers for radar grid maps which are beneficial for semantic segmentation tasks.
Semantic Segmentation on 3D Occupancy Grids for Automotive Radar
This paper uses semantic segmentation networks to distinguish between frequently occurring infrastructure objects and finds that 3D input images lead to improvements in the car park dataset, and 3D grids allow for good association with camera images.
Semantic Segmentation on Automotive Radar Maps
This paper uses popular semantic segmentation networks in order to classify the vehicle's immediate infrastructure, and shows an overall intersection over union accuracy of 95 and 90%, being a new benchmark for this young research field.
Deep Learning-based Object Classification on Automotive Radar Spectra
This work proposes to apply deep Convolutional Neural Networks (CNNs) directly to regions-of-interest (ROI) in the radar spectrum and thereby achieve an accurate classification of different objects in a scene.
CNN Based Road User Detection Using the 3D Radar Cube
This letter presents a novel radar based, single-frame, multi-class detection method for moving road users (pedestrian, cyclist, car), which utilizes low-level radar cube data and demonstrates that the method outperforms the state-of-the-art methods both target- and object-wise.