Localization Uncertainty Estimation for Anchor-Free Object Detection

Abstract

Since many safety-critical systems such as surgical robots and autonomous driving cars are in unstable environments with sensor noise or incomplete data, it is desirable for object detectors to take the confidence of the localization prediction into account. Recent attempts to estimate localization uncertainty for object detection focus only anchor-based method that captures the uncertainty of different characteristics such as location (center point) and scale (width, height). Also, anchor-based methods need to adjust sensitive anchor-box settings. Therefore, we propose a new object detector called Gaussian-FCOS that estimates the localization uncertainty based on an anchor-free detector that captures the uncertainty of similar property with four directions of box offsets (left, right, top, bottom) and avoids the anchor tuning. For this purpose, we design a new loss function, uncertainty loss, to measure how uncertain the estimated object location is by modeling the uncertainty as a Gaussian distribution. Then, the detection score is calibrated through the estimated uncertainty. Experiments on challenging COCO datasets demonstrate that the proposed new loss function not only enables the network to estimate the uncertainty but produces a synergy effect with regression loss. In addition, our Gaussian-FCOS reduces false positives with the estimated localization uncertainty and finds more missing-objects, boosting both Average Precision (AP) and Recall (AR). We hope Gaussian-FCOS serve as a baseline for the reliability-required task.Preprint. Under review.

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

“…using additional layers at the detection head to regress probability parameters, and 4) training the modified detector by incorporating uncertainty in the loss function. The direct-modelling approach has been used in various object detection architectures, including SSD [113], Faster-RCNN [8], [11], [13], FCOS [118], Point-RCNN [115], and PIXOR [10].…”

Gradient-Based Quantification of Epistemic Uncertainty for Deep Object Detectors

“…Recent progress in the estimation of aleatoric uncertainty has been made in [20,21,22,23,24,25]. In [20], a two-stage detector is adapted to learn the IoU of its prediction with the ground truth directly.…”

“…In [20], a two-stage detector is adapted to learn the IoU of its prediction with the ground truth directly. Meanwhile, approaches such as [21,22,23,24,26,27] learn the bounding box localization regression for various architectures as Gaussian distributions for each localization variable individually. These approaches aim at learning aleatoric uncertainty from the training dataset in the manner described in [10].…”

Estimating and Evaluating Regression Predictive Uncertainty in Deep Object Detectors

“…State-of-the-art probabilistic object detectors model predictive uncertainty by adapting the work of Kendall & Gal (2017) to state-of-the-art object detectors. Standard detectors are extended with a variance network, usually referred to as the variance regression head, alongside the mean bounding box regression head and the resulting network is trained using NLL (Feng et al, 2018a;Le et al, 2018;He et al, 2019;Lee et al, 2020;Feng et al, 2020;He & Wang, 2020). Some approaches combine the variance networks with dropout (Feng et al, 2018b;Kraus & Dietmayer, 2019) and use Monte Carlo sampling at test time.…”

“…This paper aims to identify the shortcomings of recent trends followed by state-of-the-art probabilistic object detectors, and proposes to provide theoretically founded solutions for identified issues. Specifically, we observe that the majority of state-of-the-art probabilistic object detectors methods (Feng et al, 2018a;Le et al, 2018;Feng et al, 2018b;He et al, 2019;Kraus & Dietmayer, 2019;Meyer et al, 2019;Choi et al, 2019;Feng et al, 2020;He & Wang, 2020;Harakeh et al, 2020;Lee et al, 2020) build on deterministic object detection backends to estimate bounding box predictive distributions by modifying such backends with variance networks (Detlefsen et al, 2019). The mean and variance of bounding box predictive distributions estimated using variance networks are then learnt using negative log likelihood (NLL).…”