Multi-Level Domain Adaptive Learning for Cross-Domain Detection

Xie, Rongchang; Yu, F. Richard; Wang, Jiachao; Wang, Yizhou; Zhang, Li

doi:10.1109/iccvw.2019.00401

Cited by 102 publications

(61 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several works have recently attempted to apply adversarial learning of UDA to the cross-domain adaptive object detection research. Current methods typically minimize the domain discrepancy at multiple levels [13]- [16]. Chen et al [17] reduce the domain discrepancy on both image and instance level.…”

Section: Introductionmentioning

confidence: 99%

Multi-Scale Feature Enhanced Domain Adaptive Object Detection For Power Transmission Line Inspection

Zhang

Zhang²,

Hao

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Domain adaptive object detection aims to build an object detector for the unlabeled target domain by transferring knowledge from a well-labeled source domain, which can alleviate the problem of cumbersome labeling of object detection in cross-scene power transmission line inspection. Remarkable advances are made recently by mitigating distributional shifts via hierarchical domain feature alignment training of detection networks. However, domain adaptive object detection is still limited in learning the invariance representation of multi-scale features. Specifically, the scale of objects varies in the scenes of aerial inspection, which hinders the knowledge transfer from the labeled source domain. In this paper, we propose a multi-scale feature enhanced domain adaptation method for cross-domain object detection of power transmission lines inspection. The proposed method consists of two components: 1) Multi-Scale Fusion Feature Alignment module (MSFA) to strengthen similar representation characteristics of different scales object in domain adaptive by utilizing context information conveyed from other levels; 2) Multi-Scale Consistency Regularization module (MSCR) to jointly optimize the multi-scale feature learning of each level, which promotes domain invariant feature learning at each level. Experimental results demonstrate that our method significantly increases the performance of the object detector in several cross-scene transmission line inspection tasks.

show abstract

Section: Introductionmentioning

confidence: 99%

Multi-Scale Feature Enhanced Domain Adaptive Object Detection For Power Transmission Line Inspection

Zhang

Zhang²,

Hao

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Unsupervised Domain Adaptation for Object Detection. Domain adaptive object detection has received much attention in the past two years [24,25,122,164,[171][172][173][174]. The DA-Faster [25] propose to align domains at both image-level and instance-level by adding two domain classifiers to the Faster R-CNN.…”

Section: Unsupervised Domain Adaptationmentioning

confidence: 99%

“…However, due to the limitation of domain adversarial training and inaccurate instance predictions, the improvement is limited. To improve the efficiency of image-level adaptation, multi-feature alignment [24,164,171,173] has been proposed. In strong-weak domain alignment (SWDA) [24], Saito et alproposed to strongly align low-level image features and weakly align highlevel image features.…”

Section: Unsupervised Domain Adaptationmentioning

confidence: 99%

“…Focal loss [43] is used in the domain classifier to achieve it. Similar works [164,171,173] aligned features from multiple layers for image-level adaptation. To address the inaccurate instance problem in instance-level adaptation, Zhu et al [122] proposed to first filter non-object instances via grouping and then emphasize the target instances that are similar to the source for adversarial domain alignment.…”

Section: Unsupervised Domain Adaptationmentioning

confidence: 99%

“…Therefore, it is possible to apply our proposed method to the tasks of semantic segmentation and instance segmentation, which might suffer from similar problems. Moreover, our proposed forward-backward cyclic method could be complementary to the other adaptation methods, such as synthesizing images via CycleGAN [166], weak alignment [24] and multi-level feature alignment [164,173].…”

Section: Chapter 8 Conclusionmentioning

confidence: 99%

See 2 more Smart Citations

On the robustness of object and face detection: false positives, attacks and adaptability

Siqi¹

View full text Add to dashboard Cite

Object detection is to recognize and locate objects in images, which is one of the fundamental problems in computer vision. It has been developed rapidly in the past decades and has a wide range of applications, e.g., surveillance systems, autonomous vehicles, medical imaging diagnosis, and industrial automation systems. Since the technique of object detection has been widely adopted, the lack of robustness of object detectors can be problematic, which may lead to unpredictable safety and financial losses. The main goal of this thesis is to study the robustness of object detectors and develop robust detectors. Robust object detectors can be viewed as the detectors that have the ability to maintain high accuracy when encountering the challenging conditions that may degrade the performance. To develop robust object detectors, we study the robustness of object detection by understanding when and where the detectors may fail. In this thesis, we focus on the failure cases in three aspects: false positives, attacks, and domain adaptability. False positives are the predictions that do not match the ground truth bounding boxes. The robustness to false positives is crucial for object detection, particularly for face detection, as face detection problem is the very first step for facial analysis tasks. The existence of false positives will waste the computations and harm the accuracy of the subsequent processes. To make the current detectors more robust, this thesis focuses on reducing the number of false positives with the detection rate well-maintained. To achieve this, we first explore different post-processing false positive classifiers to remove false positives. We further study the properties of true and false positives among face detectors and propose a framework that cascades two off-the-shelf detectors. The cascaded detector demonstrated its effectiveness and efficiency on false positive removal. Attacks are one of the biggest threats of a robust detection system, especially for surveillance systems. For example, criminals can apply attacks to the surveillance system for preventing themselves from being detected. Adversarial attacks are the attacks that can change the neural network output significantly by adding tiny, imperceptible perturbations onto the image. We examine how robust the current detectors are against adversarial attacks and provide theoretical explanations for why existing adversarial attack methods fail. We successfully perform adversarial attacks on deep learning based detection networks. Due to the variations in shape and appearance, lighting conditions and backgrounds, a model trained on the source data might not perform well on the target data, which is often known as domain discrepancy. Supervised domain adaptation methods require a large number of annotated bounding boxes for object detection, which is time-consuming and expensive. Hence, an effective model that can adapt object detectors into a new domain without labels is highly desirable. In this thesis, we propose an adaptation method t...

show abstract

Source data‐free domain adaptation of object detector through domain‐specific perturbation

Xiong

Zhang

et al. 2021

Int J Intell Syst

View full text Add to dashboard Cite

The current unsupervised cross‐domain detection methods need source domain data to retrain the detection model in target domain. However, the source domain data may be unavailable due to privacy, decentralization, or computation resource restrictions. A natural idea is to optimize the parameters of the source domain model by self‐supervised learning based on pseudo labels. We propose another approach from the viewpoint of noise perturbation without pseudo‐labeling. It can be assumed that the source and target domains are actually derived from a domain invariant space through domain‐specific perturbations, respectively. A super target domain can be constructed by augmenting more target domain perturbations to the target domain images. The optimal direction of the target domain to the domain invariant space can be approximated as the alignment direction from the super target domain to the target domain. Based on this idea, we propose a novel method called SOAP (SOurce data‐free domain Adaptation through domain Perturbation) which can remove domain perturbation from the target domain. The image‐level, instance‐level, and category consistency regularizations based on Mean Teacher structure are proposed to learn the correct alignment direction. Specifically, the category consistency can also further improve the classification accuracy. Extensive experiments on multiple domain adaptation scenarios demonstrate that SOAP achieves better performance surpassing the baseline (Faster R‐CNN) and multiple state‐of‐the‐art domain adaptation methods which need to access source domain data.

show abstract

Multi-Level Domain Adaptive Learning for Cross-Domain Detection

Cited by 102 publications

References 24 publications

Multi-Scale Feature Enhanced Domain Adaptive Object Detection For Power Transmission Line Inspection

Multi-Scale Feature Enhanced Domain Adaptive Object Detection For Power Transmission Line Inspection

On the robustness of object and face detection: false positives, attacks and adaptability

Source data‐free domain adaptation of object detector through domain‐specific perturbation

Contact Info

Product

Resources

About