BayesOD: A Bayesian Approach for Uncertainty Estimation in Deep Object Detectors

Harakeh, Ali; Smart, Michael; Waslander, Steven L.

doi:10.1109/icra40945.2020.9196544

Cited by 97 publications

(88 citation statements)

References 21 publications

(48 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the promising role of uncertainty estimation in increasing autonomous and robotic systems' safety by indicating low confidence in output predictions -and consequently detecting failures -many authors in the field of robotic perception investigated and compared variations of the main methods of estimating uncertainty from DNNs. Examples include uncertainty estimation for steering angle estimation [43], road segmentation [44], visual odometry [45], and vehicle and object detection [46]- [48].…”

Section: ) Uncertainty Estimationmentioning

confidence: 99%

Run-Time Monitoring of Machine Learning for Robotic Perception: A Survey of Emerging Trends

2021

View full text Add to dashboard Cite

As deep learning continues to dominate all state-of-the-art computer vision tasks, it is increasingly becoming an essential building block for robotic perception. This raises important questions concerning the safety and reliability of learning-based perception systems. There is an established field that studies safety certification and convergence guarantees of complex software systems at design-time. However, the unknown future deployment environments of an autonomous system and the complexity of learning-based perception make the generalization of design-time verification to run-time problematic. In the face of this challenge, more attention is starting to focus on run-time monitoring of performance and reliability of perception systems with several trends emerging in the literature in the face of this challenge. This paper attempts to identify these trends and summarise the various approaches to the topic.

show abstract

Section: ) Uncertainty Estimationmentioning

confidence: 99%

Run-Time Monitoring of Machine Learning for Robotic Perception: A Survey of Emerging Trends

2021

View full text Add to dashboard Cite

show abstract

“…In recent years, Dropout Variational Inference [27] has made Bayesian Neural Networks (BNN) a tractable solution to provide epistemic uncertainty quantifications for deep neural networks, and Miller et al [8] applied this method to object detection for the first time. Currently, increasing attention [1,9,10,28] has been devoted to quantify uncertainties in the results of object detectors.…”

Section: Online Performance Predictionmentioning

confidence: 99%

“…To mitigate this problem, some prior works [ 1 , 8 , 9 , 10 ] are devoted to outputting the uncertainty information of the detection results during the online inference process, but these methods can only provide uncertainties for detected objects in the outputs, and cannot deal with FNs. Some studies [ 4 , 11 ] try to predict online performance metrics, such as Average Precision (AP), of an object detector; however, these predicted metrics can only reflect the overall detection performance on the input image, and cannot provide specific object-wise failure predictions, which we believe are more indispensable and constructive for the application of autonomous driving.…”

Section: Introductionmentioning

confidence: 99%

Introspective False Negative Prediction for Black-Box Object Detectors in Autonomous Driving

Yang

Chen

et al. 2021

Sensors

View full text Add to dashboard Cite

Object detection plays a critical role in autonomous driving, but current state-of-the-art object detectors will inevitably fail in many driving scenes, which is unacceptable for safety-critical automated vehicles. Given the complexity of the real traffic scenarios, it is impractical to guarantee zero detection failure; thus, online failure prediction is of crucial importance to mitigate the risk of traffic accidents. Of all the failure cases, False Negative (FN) objects are most likely to cause catastrophic consequences, but little attention has been paid to the online FN prediction. In this paper, we propose a general introspection framework that can make online prediction of FN objects for black-box object detectors. In contrast to existing methods which rely on empirical assumptions or handcrafted features, we facilitate the FN feature extraction by an introspective FN predictor we designed in this framework. For this purpose, we extend the original concept of introspection to object-wise FN predictions, and propose a multi-branch cooperation mechanism to address the distinct foreground-background imbalance problem of FN objects. The effectiveness of the proposed framework is verified through extensive experiments and analysis, and the results show that our method successfully predicts the FN objects with 81.95% precision for 88.10% recall on the challenging KITTI Benchmark, and effectively improves object detection performance by taking FN predictions into consideration.

show abstract

“…Uncertainty techniques for detection tasks [21,22] also involve a combination of sampling-based methods [23][24][25] and non-sampling-based methods [26][27][28]. Miller et al evaluate the performance of object detectors in open-set conditions.…”

Section: Uncertainty Estimation For Detection Tasksmentioning

confidence: 99%

A Task-Driven Feedback Imager with Uncertainty Driven Hybrid Control

Mudassar

Saha

Wolf

et al. 2021

Sensors

View full text Add to dashboard Cite

Deep Neural Network (DNN) systems tend to produce overconfident or uncalibrated outputs. This poses problems for active sensor systems that have a DNN module as the main feedback controller. In this paper, we study a closed-loop feedback smart camera from the lens of uncertainty estimation. The uncertainty of the task output is used to characterize and facilitate the feedback operation. The DNN uncertainty in the feedback system is estimated and characterized using both sampling and non-sampling based methods. In addition, we propose a closed-loop control that incorporates uncertainty information when providing feedback. We show two modes of control, one that prioritizes false positives and one that prioritizes false negatives, and a hybrid approach combining the two. We apply the uncertainty-driven control to the tasks of object detection, object tracking, and action detection. The hybrid system improves object detection and tracking accuracy on the CAMEL dataset by 1.1% each respectively. For the action detection task, the hybrid approach improves accuracy by 1.4%.

show abstract

BayesOD: A Bayesian Approach for Uncertainty Estimation in Deep Object Detectors

Cited by 97 publications

References 21 publications

Run-Time Monitoring of Machine Learning for Robotic Perception: A Survey of Emerging Trends

Run-Time Monitoring of Machine Learning for Robotic Perception: A Survey of Emerging Trends

Introspective False Negative Prediction for Black-Box Object Detectors in Autonomous Driving

A Task-Driven Feedback Imager with Uncertainty Driven Hybrid Control

Contact Info

Product

Resources

About