A Review and Comparative Study on Probabilistic Object Detection in Autonomous Driving

Feng, Di; Harakeh, Ali; Waslander, Steven; Dietmayer, Klaus

doi:10.48550/arxiv.2011.10671

Cited by 6 publications

(8 citation statements)

References 95 publications

(236 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Faster R-CNN [38], a two-stage object detector. We use the Faster R-CNN and RetinaNet implementations available at [39] and [27] respectively. For both detectors, we use a ResNet-50 and feature pyramid network (FPN) backbone.…”

Section: Comparison Detectors and State-of-the-art Methodsmentioning

confidence: 99%

“…Epistemic uncertainty is uncertainty in a model's parameters due to lack of knowledge or data [5], and can be used to identify inputs that are not reflected by the model's training data -epistemic uncertainty is required for identifying open-set errors [26]. We refer the reader to [27] for more information on uncertainty techniques in object detection.…”

Section: Related Work a Uncertainty Estimation In Object Detectionmentioning

confidence: 99%

See 1 more Smart Citation

Uncertainty for Identifying Open-Set Errors in Visual Object Detection

Miller,

Sünderhauf,

Milford

et al. 2021

Preprint

View full text Add to dashboard Cite

Deployed into an open world, object detectors are prone to a type of false positive detection termed open-set errors. We propose GMM-Det, a real-time method for extracting epistemic uncertainty from object detectors to identify and reject open-set errors. GMM-Det trains the detector to produce a structured logit space that is modelled with class-specific Gaussian Mixture Models. At test time, open-set errors are identified by their low log-probability under all Gaussian Mixture Models. We test two common detector architectures, Faster R-CNN and RetinaNet, across three varied datasets spanning robotics and computer vision. Our results show that GMM-Det consistently outperforms existing uncertainty techniques for identifying and rejecting open-set detections, especially at the low-error-rate operating point required for safety-critical applications. GMM-Det maintains object detection performance, and introduces only minimal computational overhead. We also introduce a methodology for converting existing object detection datasets into specific open-set datasets to consistently evaluate open-set performance in object detection.

show abstract

Section: Comparison Detectors and State-of-the-art Methodsmentioning

confidence: 99%

Section: Related Work a Uncertainty Estimation In Object Detectionmentioning

confidence: 99%

Uncertainty for Identifying Open-Set Errors in Visual Object Detection

Miller,

Sünderhauf,

Milford

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…We consider different ways to apply DropBlock in YOLO models and show how it can improve the uncertainty modeling, generalization performance, and calibration capabilities of the YOLO models. We compare the proposed MC DropBlock with various baselines including the ones considering drop block only at training time (training time dropBlock) or only at inference or test time (inference time dropBlock), and with existing approaches such as Dropout based [37] and Gaussian Yolo models [38]. All the experiments are trained with NVIDIA V100 GPUs.…”

Section: Methodsmentioning

confidence: 99%

Monte Carlo DropBlock for Modelling Uncertainty in Object Detection

Deepshikha,

Yelleni,

Srijith

et al. 2021

Preprint

View full text Add to dashboard Cite

With the advancements made in deep learning, computer vision problems like object detection and segmentation have seen a great improvement in performance. However, in many real-world applications such as autonomous driving vehicles, the risk associated with incorrect predictions of objects is very high. Standard deep learning models for object detection such as YOLO models are often overconfident in their predictions and do not take into account the uncertainty in predictions on out-of-distribution data. In this work, we propose an efficient and effective approach to model uncertainty in object detection and segmentation tasks using Monte-Carlo DropBlock (MC-DropBlock) based inference. The proposed approach applies drop-block during training time and test time on the convolutional layer of the deep learning models such as YOLO. We show that this leads to a Bayesian convolutional neural network capable of capturing the epistemic uncertainty in the model. Additionally, we capture the aleatoric uncertainty using a Gaussian likelihood. We demonstrate the effectiveness of the proposed approach on modeling uncertainty in object detection and segmentation tasks using out-of-distribution experiments. Experimental results show that MC-DropBlock improves the generalization, calibration, and uncertainty modeling capabilities of YOLO models in object detection and segmentation.

show abstract

“…Along with its theoretic soundness, the task-agnostic characteristic of these approaches can facilitate wider applicability for different domains. Most of previous work integrate BNNs into the detection head of current object detectors [35]. While some only exploit the classification branch for the uncertainty estimation [36], others [37], [38] consider both classification and regression branches.…”

Section: Related Work A)mentioning

confidence: 99%

Bridging the Last Mile in Sim-to-Real Robot Perception via Bayesian Active Learning

Feng¹,

Lee²,

Durner³

et al. 2021

Preprint

View full text Add to dashboard Cite

Learning from synthetic data is popular in a variety of robotic vision tasks such as object detection, because large amount of data can be generated without annotations by humans. However, when relying only on synthetic data, we encounter the well-known problem of the simulation-toreality (Sim-to-Real) gap, which is hard to resolve completely in practice. For such cases, real human-annotated data is necessary to bridge this gap, and in our work we focus on how to acquire this data efficiently. Therefore, we propose a Simto-Real pipeline that relies on deep Bayesian active learning and aims to minimize the manual annotation efforts. We devise a learning paradigm that autonomously selects the data that is considered useful for the human expert to annotate. To achieve this, a Bayesian Neural Network (BNN) object detector providing reliable uncertain estimates is adapted to infer the informativeness of the unlabeled data, in order to perform active learning. In our experiments on two object detection data sets, we show that the labeling effort required to bridge the reality gap can be reduced to a small amount. Furthermore, we demonstrate the practical effectiveness of this idea in a grasping task on an assistive robot.

show abstract

A Review and Comparative Study on Probabilistic Object Detection in Autonomous Driving

Cited by 6 publications

References 95 publications

Uncertainty for Identifying Open-Set Errors in Visual Object Detection

Uncertainty for Identifying Open-Set Errors in Visual Object Detection

Monte Carlo DropBlock for Modelling Uncertainty in Object Detection

Bridging the Last Mile in Sim-to-Real Robot Perception via Bayesian Active Learning

Contact Info

Product

Resources

About