Towards Safe Autonomous Driving: Capture Uncertainty in the Deep Neural Network For Lidar 3D Vehicle Detection

By Di Feng, Lars Rosenbaum, Klaus C. J. Dietmayer. 2018

283 citations recorded by the source corpus.

To assure that an autonomous car is driving safely on public roads, its object detection module should not only work correctly, but show its prediction confidence as well. Previous object detectors driven by deep learning do not explicitly model uncertainties in the neural network. We tackle with this problem by presenting practical methods to capture uncertainties in a 3D vehicle detector for Lidar point clouds. The proposed probabilistic detector represents reliable epistemic uncertainty and aleatoric uncertainty in classification and localization tasks. Experimental results show that the epistemic uncertainty is related to the detection accuracy, whereas the aleatoric uncertainty is influenced by vehicle distance and occlusion. The results also show that we can improve the detection performance by 1%–5% by modeling the aleatoric uncertainty.

Subfields: Automotive Engineering, Computer Vision and Pattern Recognition, Environmental Engineering, 3D Vision & Pose, Robotics & Control

How researchers see this paper

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

In Automotive Engineering

In Computer Vision and Pattern Recognition

In Environmental Engineering

In 3D Vision & Pose

In Robotics & Control

Citations in the corpus

Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awareness (2020, methodology): “Efficient methods that reliably quantify a deep neural network (DNN)’s predictive uncertainty are important for industrial-scale, real-world applications, which include examples such as object recognition in autonomous driving [21], ad click prediction in online advertising [74], and intent understa”
A General Framework for Uncertainty Estimation in Deep Learning (2019, methodology): “Given the paramount importance of safety, autonomous driving research has allocated a lot of attention to the problem of uncertainty estimation, from both the perception [18], [19] and the control side [13], [14].”
A Survey of Autonomous Driving: Common Practices and Emerging Technologies (2019, background): “Several other works have reused camera-based algorithms and trained efficient networks for 3D object detection on 2D BV images [170]–[173].”
A Survey on 3D Object Detection Methods for Autonomous Driving Applications (2019, methodology): “[53] uses a Bayesian Neural Network to predict the class and 3D bounding box after ROI pooling, which allows to quantify the network confidence for both outputs.”
Gaussian YOLOv3: An Accurate and Fast Object Detector Using Localization Uncertainty for Autonomous Driving (2019, methodology): “[6] proposed a method for predicting uncertainty by applying Kendall et al.”
A Review and Comparative Study on Probabilistic Object Detection in Autonomous Driving (2020, methodology): “For example, Feng et al .”
Deep Multi-Modal Object Detection and Semantic Segmentation for Autonomous Driving: Datasets, Methods, and Challenges (2019, background): “[212] examine the epistemic and aleatoric uncertainties in a LiDAR vehicle detection network for autonomous driving.”
MonoPair: Monocular 3D Object Detection Using Pairwise Spatial Relationships (2020, methodology): “This concept has been applied in 3D Lidar object detection [12] and pedestrian localization [2], where they mainly consider uncertainties as additional information for reference.”
Evaluating Merging Strategies for Sampling-based Uncertainty Techniques in Object Detection (2018, background): “This has only recently been demonstrated for the first time for object detection under open-set conditions in [8], and for vehicle detection in 3D Lidar data by [23].”
LaserNet: An Efficient Probabilistic 3D Object Detector for Autonomous Driving (2019, unknown): “Based on this work, [6, 7] produce and analyze epistemic and aleatoric uncertainties in 3D object detection.”
A survey of uncertainty in deep neural networks (2021, background): “medical image analysis [1], [2], [3], [4], [5], [6] or autonomous vehicle control [7], [8], [9], [10], their deployment in mission-”
Bounding Box Regression With Uncertainty for Accurate Object Detection (2018, background): “The idea is further extended to the 3D object detection [10, 9].”