Cross-domain Learning Using Optimized Pseudo Labels: Towards Adaptive Car Detection in Different Weather Conditions and Urban Cities

2021 ◽  
Author(s):  
ke wang ◽  
Lianhua Zhang ◽  
Qin Xia ◽  
Liang Pu ◽  
Junlan Chen
Keyword(s):  
Autonomous Vehicles ◽  
Adaptive Sampling ◽  
Domain Adaptation ◽  
Weather Conditions ◽  
Training Data ◽  
Target Domain ◽  
Average Precision ◽  
Sharp Drop ◽  
Car Detection ◽  
Hard Samples

Abstract Convolutional neural networks (CNN) based object detection usually assumes that training and test data have the same distribution, which, however, does not always hold in real-world applications. In autonomous vehicles, the driving scene (target domain) consists of unconstrained road environments which cannot all possibly be observed in training data (source domain) and this will lead to a sharp drop in the accuracy of the detector. In this paper, we propose a domain adaptation framework based on pseudo-labels to solve the domain shift. First, the pseudo-labels of the target domain images are generated by the baseline detector (BD) and optimized by our data optimization module to correct the errors. Then, the hard samples in a single image are labeled based on the optimization results of pseudo-labels. The adaptive sampling module is approached to sample target domain data according to the number of hard samples per image to select more effective data. Finally, a modified knowledge distillation loss is applied in the retraining module, and we investigate two ways of assigning soft-labels to the training examples from the target domain to retrain the detector. We evaluate the average precision of our approach in various source/target domain pairs and demonstrate that the framework improves over 10% average precision of BD on multiple domain adaptation scenarios on the Cityscapes, KITTI, and Apollo datasets.

scholarly journals Lightweight Object Detection Ensemble Framework for Autonomous Vehicles in Challenging Weather Conditions

2021 ◽  
Vol 2021 ◽  
pp. 1-12
Author(s):  
Rahee Walambe ◽  
Aboli Marathe ◽  
Ketan Kotecha ◽  
George Ghinea
Keyword(s):  
Computer Vision ◽  
Object Detection ◽  
Autonomous Vehicles ◽  
Data Augmentation ◽  
Weather Conditions ◽  
Training Data ◽  
Average Precision ◽  
Ensemble Techniques ◽  
Rainy Weather ◽  
Increase In Accuracy

The computer vision systems driving autonomous vehicles are judged by their ability to detect objects and obstacles in the vicinity of the vehicle in diverse environments. Enhancing this ability of a self-driving car to distinguish between the elements of its environment under adverse conditions is an important challenge in computer vision. For example, poor weather conditions like fog and rain lead to image corruption which can cause a drastic drop in object detection (OD) performance. The primary navigation of autonomous vehicles depends on the effectiveness of the image processing techniques applied to the data collected from various visual sensors. Therefore, it is essential to develop the capability to detect objects like vehicles and pedestrians under challenging conditions such as like unpleasant weather. Ensembling multiple baseline deep learning models under different voting strategies for object detection and utilizing data augmentation to boost the models’ performance is proposed to solve this problem. The data augmentation technique is particularly useful and works with limited training data for OD applications. Furthermore, using the baseline models significantly speeds up the OD process as compared to the custom models due to transfer learning. Therefore, the ensembling approach can be highly effective in resource-constrained devices deployed for autonomous vehicles in uncertain weather conditions. The applied techniques demonstrated an increase in accuracy over the baseline models and were able to identify objects from the images captured in the adverse foggy and rainy weather conditions. The applied techniques demonstrated an increase in accuracy over the baseline models and reached 32.75% mean average precision (mAP) and 52.56% average precision (AP) in detecting cars in the adverse fog and rain weather conditions present in the dataset. The effectiveness of multiple voting strategies for bounding box predictions on the dataset is also demonstrated. These strategies help increase the explainability of object detection in autonomous systems and improve the performance of the ensemble techniques over the baseline models.

An intelligent fault diagnosis method based on domain adaptation for rolling bearings under variable load conditions

2021 ◽  
pp. 095440622110329
Author(s):  
Jianqun Zhang ◽  
Qing Zhang ◽  
Xianrong Qin ◽  
Yuantao Sun
Keyword(s):  
Feature Extraction ◽  
Fault Diagnosis ◽  
Domain Adaptation ◽  
Rolling Bearing ◽  
Training Data ◽  
Variable Load ◽  
K Nearest Neighbor ◽  
Target Domain ◽  
Bearing Faults ◽  
Load Conditions

To identify rolling bearing faults under variable load conditions, a method named DISA-KNN is proposed in this paper, which is based on the strategy of feature extraction-domain adaptation-classification. To be specific, the time-domain and frequency-domain indicators are used for feature extraction. Discriminative and domain invariant subspace alignment (DISA) is used to minimize the data distributions’ discrepancies between the training data (source domain) and testing data (target domain). K-nearest neighbor (KNN) is applied to identify rolling bearing faults. DISA-KNN’s validation is proved by the experimental signal collected under different load conditions. The identification accuracies obtained by the DISA-KNN method are more than 90% on four datasets, including one dataset with 99.5% accuracy. The strength of the proposed method is further highlighted by comparisons with the other 8 methods. These results reveal that the proposed method is promising for the rolling bearing fault diagnosis in real rotating machinery.

scholarly journals USING SEMANTICALLY PAIRED IMAGES TO IMPROVE DOMAIN ADAPTATION FOR THE SEMANTIC SEGMENTATION OF AERIAL IMAGES

2020 ◽  
Vol V-2-2020 ◽  
pp. 483-492
Author(s):  
D. Gritzner ◽  
J. Ostermann
Keyword(s):  
Time Window ◽  
Domain Adaptation ◽  
Geographical Area ◽  
Model Performance ◽  
Ground Truth ◽  
Semantic Segmentation ◽  
Training Data ◽  
Aerial Images ◽  
Target Domain ◽  
Training Examples

Abstract. Modern machine learning, especially deep learning, which is used in a variety of applications, requires a lot of labelled data for model training. Having an insufficient amount of training examples leads to models which do not generalize well to new input instances. This is a particular significant problem for tasks involving aerial images: often training data is only available for a limited geographical area and a narrow time window, thus leading to models which perform poorly in different regions, at different times of day, or during different seasons. Domain adaptation can mitigate this issue by using labelled source domain training examples and unlabeled target domain images to train a model which performs well on both domains. Modern adversarial domain adaptation approaches use unpaired data. We propose using pairs of semantically similar images, i.e., whose segmentations are accurate predictions of each other, for improved model performance. In this paper we show that, as an upper limit based on ground truth, using semantically paired aerial images during training almost always increases model performance with an average improvement of 4.2% accuracy and .036 mean intersection-over-union (mIoU). Using a practical estimate of semantic similarity, we still achieve improvements in more than half of all cases, with average improvements of 2.5% accuracy and .017 mIoU in those cases.

scholarly journals Multi-Source Distilling Domain Adaptation

2020 ◽  
Vol 34 (07) ◽  
pp. 12975-12983
Author(s):  
Sicheng Zhao ◽  
Guangzhi Wang ◽  
Shanghang Zhang ◽  
Yang Gu ◽  
Yaxian Li ◽  
...  
Keyword(s):  
Domain Adaptation ◽  
Feature Space ◽  
Wasserstein Distance ◽  
Training Data ◽  
Source Distribution ◽  
Single Source ◽  
Multiple Sources ◽  
Target Domain ◽  
Target Feature ◽  
Training Samples

Deep neural networks suffer from performance decay when there is domain shift between the labeled source domain and unlabeled target domain, which motivates the research on domain adaptation (DA). Conventional DA methods usually assume that the labeled data is sampled from a single source distribution. However, in practice, labeled data may be collected from multiple sources, while naive application of the single-source DA algorithms may lead to suboptimal solutions. In this paper, we propose a novel multi-source distilling domain adaptation (MDDA) network, which not only considers the different distances among multiple sources and the target, but also investigates the different similarities of the source samples to the target ones. Specifically, the proposed MDDA includes four stages: (1) pre-train the source classifiers separately using the training data from each source; (2) adversarially map the target into the feature space of each source respectively by minimizing the empirical Wasserstein distance between source and target; (3) select the source training samples that are closer to the target to fine-tune the source classifiers; and (4) classify each encoded target feature by corresponding source classifier, and aggregate different predictions using respective domain weight, which corresponds to the discrepancy between each source and target. Extensive experiments are conducted on public DA benchmarks, and the results demonstrate that the proposed MDDA significantly outperforms the state-of-the-art approaches. Our source code is released at: https://github.com/daoyuan98/MDDA.

scholarly journals Distributional Correspondence Indexing for Cross-Lingual and Cross-Domain Sentiment Classification (Extended Abstract)

2018 ◽  
Author(s):  
Alejandro Moreo Fernández ◽  
Andrea Esuli ◽  
Fabrizio Sebastiani
Keyword(s):  
Domain Adaptation ◽  
State Of The Art ◽  
Sentiment Classification ◽  
Training Data ◽  
Target Domain ◽  
Source Domain ◽  
Machine Learning Methods ◽  
Cross Domain ◽  
Current State ◽  
Cross Lingual

Domain Adaptation (DA) techniques aim at enabling machine learning methods learn effective classifiers for a “target” domain when the only available training data belongs to a different “source” domain. In this extended abstract, we briefly describe our new DA method called Distributional Correspondence Indexing (DCI) for sentiment classification. DCI derives term representations in a vector space common to both domains where each dimension reflects its distributional correspondence to a pivot, i.e., to a highly predictive term that behaves similarly across domains. The experiments we have conducted show that DCI obtains better performance than current state-of-the-art techniques for cross-lingual and cross-domain sentiment classification.

scholarly journals ITERATIVE RE-WEIGHTED INSTANCE TRANSFER FOR DOMAIN ADAPTATION

2016 ◽  
Vol III-3 ◽  
pp. 339-346 ◽  
Author(s):  
A. Paul ◽  
F. Rottensteiner ◽  
C. Heipke
Keyword(s):  
Domain Adaptation ◽  
Training Data ◽  
Aerial Images ◽  
Target Domain ◽  
Data Set ◽  
Source Domain ◽  
The Difference ◽  
A New Technique ◽  
Class Labels ◽  
Decision Boundaries

Domain adaptation techniques in transfer learning try to reduce the amount of training data required for classification by adapting a classifier trained on samples from a source domain to a new data set (target domain) where the features may have different distributions. In this paper, we propose a new technique for domain adaptation based on logistic regression. Starting with a classifier trained on training data from the source domain, we iteratively include target domain samples for which class labels have been obtained from the current state of the classifier, while at the same time removing source domain samples. In each iteration the classifier is re-trained, so that the decision boundaries are slowly transferred to the distribution of the target features. To make the transfer procedure more robust we introduce weights as a function of distance from the decision boundary and a new way of regularisation. Our methodology is evaluated using a benchmark data set consisting of aerial images and digital surface models. The experimental results show that in the majority of cases our domain adaptation approach can lead to an improvement of the classification accuracy without additional training data, but also indicate remaining problems if the difference in the feature distributions becomes too large.

scholarly journals ADVERSARIAL DOMAIN ADAPTATION FOR THE CLASSIFICATION OF AERIAL IMAGES AND HEIGHT DATA USING CONVOLUTIONAL NEURAL NETWORKS

2019 ◽  
Vol IV-2/W7 ◽  
pp. 197-204 ◽  
Author(s):  
D. Wittich ◽  
F. Rottensteiner
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Domain Adaptation ◽  
Feature Representation ◽  
Training Data ◽  
Aerial Images ◽  
Remotely Sensed Data ◽  
Target Domain ◽  
Deep Convolutional Neural Networks

<p><strong>Abstract.</strong> Domain adaptation (DA) can drastically decrease the amount of training data needed to obtain good classification models by leveraging available data from a source domain for the classification of a new (target) domains. In this paper, we address deep DA, i.e. DA with deep convolutional neural networks (CNN), a problem that has not been addressed frequently in remote sensing. We present a new method for semi-supervised DA for the task of pixel-based classification by a CNN. After proposing an encoder-decoder-based fully convolutional neural network (FCN), we adapt a method for adversarial discriminative DA to be applicable to the pixel-based classification of remotely sensed data based on this network. It tries to learn a feature representation that is domain invariant; domain-invariance is measured by a classifier’s incapability of predicting from which domain a sample was generated. We evaluate our FCN on the ISPRS labelling challenge, showing that it is close to the best-performing models. DA is evaluated on the basis of three domains. We compare different network configurations and perform the representation transfer at different layers of the network. We show that when using a proper layer for adaptation, our method achieves a positive transfer and thus an improved classification accuracy in the target domain for all evaluated combinations of source and target domains.</p>

scholarly journals Fuzzy Graph Learning Regularized Sparse Filtering for Visual Domain Adaptation

2021 ◽  
Vol 11 (10) ◽  
pp. 4503
Author(s):  
Lingtong Min ◽  
Deyun Zhou ◽  
Xiaoyang Li ◽  
Qinyi Lv ◽  
Yuanjie Zhi
Keyword(s):  
Domain Adaptation ◽  
Weather Conditions ◽  
Optimization Method ◽  
Superior Performance ◽  
Target Domain ◽  
Graph Learning ◽  
Learning Framework ◽  
Cross Domain ◽  
Sparse Filtering ◽  
Visual Domain

Distribution mismatch can be easily found in multi-sensor systems, which may be caused by different shoot angles, weather conditions and so on. Domain adaptation aims to build robust classifiers using the knowledge from a well-labeled source domain, while applied on a related but different target domain. Pseudo labeling is a prevalent technique for class-wise distribution alignment. Therefore, numerous efforts have been spent on alleviating the issue of mislabeling. In this paper, unlike existing selective hard labeling works, we propose a fuzzy labeling based graph learning framework for matching conditional distribution. Specifically, we construct the cross-domain affinity graph by considering the fuzzy label matrix of target samples. In order to solve the problem of representation shrinkage, the paradigm of sparse filtering is introduced. Finally, a unified optimization method based on gradient descent is proposed. Extensive experiments show that our method achieves comparable or superior performance when compared to state-of-the-art works.

Tackling challenges of neural purchase stage identification from imbalanced twitter data

2019 ◽  
Vol 26 (4) ◽  
pp. 383-411
Author(s):  
Heike Adel ◽  
Francine Chen ◽  
Yan-Ying Chen
Keyword(s):  
Domain Adaptation ◽  
Source Model ◽  
Training Data ◽  
Fine Tuning ◽  
Target Domain ◽  
Product Domain ◽  
Main Challenge ◽  
Social Media Platforms ◽  
Aida Model ◽  
Label Distribution

AbstractTwitter and other social media platforms are often used for sharing interest in products. The identification of purchase decision stages, such as in the AIDA model (Awareness, Interest, Desire, and Action), can enable more personalized e-commerce services and a finer-grained targeting of advertisements than predicting purchase intent only. In this paper, we propose and analyze neural models for identifying the purchase stage of single tweets in a user’s tweet sequence. In particular, we identify three challenges of purchase stage identification: imbalanced label distribution with a high number of non-purchase-stage instances, limited amount of training data, and domain adaptation with no or only little target domain data. Our experiments reveal that the imbalanced label distribution is the main challenge for our models. We address it with ranking loss and perform detailed investigations of the performance of our models on the different output classes. In order to improve the generalization of the models and augment the limited amount of training data, we examine the use of sentiment analysis as a complementary, secondary task in a multitask framework. For applying our models to tweets from another product domain, we consider two scenarios: for the first scenario without any labeled data in the target product domain, we show that learning domain-invariant representations with adversarial training is most promising, while for the second scenario with a small number of labeled target examples, fine-tuning the source model weights performs best. Finally, we conduct several analyses, including extracting attention weights and representative phrases for the different purchase stages. The results suggest that the model is learning features indicative of purchase stages and that the confusion errors are sensible.

scholarly journals Unsupervised Adversarial Domain Adaptation with Error-Correcting Boundaries and Feature Adaption Metric for Remote-Sensing Scene Classification

2021 ◽  
Vol 13 (7) ◽  
pp. 1270
Author(s):  
Chenhui Ma ◽  
Dexuan Sha ◽  
Xiaodong Mu
Keyword(s):  
Remote Sensing ◽  
Domain Adaptation ◽  
Training Data ◽  
Scene Classification ◽  
Target Domain ◽  
Domain Specific ◽  
Invariant Features ◽  
Distribution Matching ◽  
Feature Adaptation ◽  
Public Datasets

Unsupervised domain adaptation (UDA) based on adversarial learning for remote-sensing scene classification has become a research hotspot because of the need to alleviating the lack of annotated training data. Existing methods train classifiers according to their ability to distinguish features from source or target domains. However, they suffer from the following two limitations: (1) the classifier is trained on source samples and forms a source-domain-specific boundary, which ignores features from the target domain and (2) semantically meaningful features are merely built from the adversary of a generator and a discriminator, which ignore selecting the domain invariant features. These issues limit the distribution matching performance of source and target domains, since each domain has its distinctive characteristic. To resolve these issues, we propose a framework with error-correcting boundaries and feature adaptation metric. Specifically, we design an error-correcting boundaries mechanism to build target-domain-specific classifier boundaries via multi-classifiers and error-correcting discrepancy loss, which significantly distinguish target samples and reduce their distinguished uncertainty. Then, we employ a feature adaptation metric structure to enhance the adaptation of ambiguous features via shallow layers of the backbone convolutional neural network and alignment loss, which automatically learns domain invariant features. The experimental results on four public datasets outperform other UDA methods of remote-sensing scene classification.

Sign in / Sign up

Export Citation Format

Share Document