LaserNet: An Efficient Probabilistic 3D Object Detector for Autonomous Driving

By Gregory P. Meyer, Ankit Laddha, Eric Kee, Carlos Vallespi-Gonzalez, Carl K. Wellington. Computer Vision and Pattern Recognition, 2019

386 citations recorded by the source corpus.

In this paper, we present LaserNet, a computationally efficient method for 3D object detection from LiDAR data for autonomous driving. The efficiency results from processing LiDAR data in the native range view of the sensor, where the input data is naturally compact. Operating in the range view involves well known challenges for learning, including occlusion and scale variation, but it also provides contextual information based on how the sensor data was captured. Our approach uses a fully convolutional network to predict a multimodal distribution over 3D boxes for each point and then it efficiently fuses these distributions to generate a prediction for each object. Experiments show that modeling each detection as a distribution rather than a single deterministic box leads to better overall detection performance. Benchmark results show that this approach has significantly lower runtime than other recent detectors and that it achieves state-of-the-art performance when compared on a large dataset that has enough data to overcome the challenges of training on the range view.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/e6c278ea0e66a0b62b5a0fa6298d2e7fbd9ec8bf

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

PointPainting: Sequential Fusion for 3D Object Detection (2019, methodology): “As a result, bird’s-eye view methods outperform top range-view methods, such as LaserNet [16, 15] on the KITTI leaderboard.”
Pseudo-LiDAR++: Accurate Depth for 3D Object Detection in Autonomous Driving (2019, methodology): “Images have also been used, but mainly to supplement LiDAR (Meyer et al., 2019a; Xu et al., 2018; Liang et al., 2018; Chen et al., 2017; Ku et al., 2018).”
Geometry Uncertainty Projection Network for Monocular 3D Object Detection (2021, neutral): “Compared with the LiDAR/stereo-based methods [32, 35, 37, 40, 41, 49, 57], monocular 3D object detection is still a challenging task due to the lack of depth cues, which makes monocular object-level depth estimation naturally ill-posed.”
Deep Multi-Modal Object Detection and Semantic Segmentation for Autonomous Driving: Datasets, Methods, and Challenges (2019, background): “Other works that introduce aleatoric uncertainties in object detectors include [214]–[217].”
Embracing Single Stride 3D Object Detector with Sparse Transformer (2021, background): “As explored in several recent works [1, 3, 13, 21, 33], range view enjoys computational advantages over voxels, especially for long-range LiDAR sensors.”
PV-RCNN++: Point-Voxel Feature Set Abstraction With Local Vector Representation for 3D Object Detection (2021, neutral): “7) Overall 0-30m 30-50m 50m-Inf Overall 0-30m 30-50m 50m-Inf LaserNet [89] 55.”
TransFuser: Imitation With Transformer-Based Sensor Fusion for Autonomous Driving (2022, background): “They operate primarily based on geometric feature projections between the image space and different LiDAR projection spaces, e.g. Bird’s Eye View (BEV) [14], [15], [16], [17], [18], [19], [20], [21], [22] and Range View (RV) [14], [17], [23], [30], [34], [35].”
DeepFusion: Lidar-Camera Deep Fusion for Multi-Modal 3D Object Detection (2022, background): “[1, 19] directly work on the range images to predict 3D bounding boxes.”
Deep Learning for Image and Point Cloud Fusion in Autonomous Driving: A Review (2020, neutral): “This feature map is fed into the LaserNet to predict per-point distributions of the bounding box and combine them for final 3D proposals.”
StarNet: Targeted Computation for Object Detection in Point Clouds (2019, methodology): “By voxelizing a point cloud (i.e. identifying a grid location for individual points (cid:126)x i ) into a series of stacked image slices describing occupancy, one may employ traditional CNN techniques for object detection on the resulting images or voxel grids [47, 53, 45, 28, 46, 52, 29].”
End-to-End Multi-View Fusion for 3D Object Detection in LiDAR Point Clouds (2019, methodology): “Recently, there has been a lot of progress on utilizing the perspective range-image, a more native representation of the raw LiDAR data (e.g., LaserNet [4]).”
Multi-Modal Fusion Transformer for End-to-End Autonomous Driving (2021, background): “Bird’s Eye View (BEV) [22, 12, 66, 9, 44, 31, 34, 61, 33] and Range View (RV) [39, 37, 22, 38, 9, 51].”
A Review and Comparative Study on Probabilistic Object Detection in Autonomous Driving (2020, methodology): “Most studies to date employ LiDARs [7]–[10], [14]–[16], [105], [115] and RGB cameras [5], [6], [12], [13], [45], [107]– [109], [113], [127] in probabilistic object detection.”
MP3: A Unified Model to Map, Perceive, Predict and Plan (2021, background): “Perception and Prediction: Most previous works perform object detection [15, 25, 34, 53, 57] and actor-based prediction to reason about the current and future state of a driving scene.”
Point2Seq: Detecting 3D Objects as Sequences (2022, methodology): “The range-based backbones [1, 8, 14, 20, 31] take range images, which are the raw data from LiDAR sensors, as the input representation.”
EPNet: Enhancing Point Features with Image Semantics for 3D Object Detection (2020, methodology): “Many LiDAR-based methods [39,24,40] are proposed in recent years.”