Cross-domain Adaptation Research Articles

RetVes segmentation: A pseudo-labeling and feature knowledge distillation optimization technique for retinal vessel channel enhancement

Recent advancements in retinal vessel segmentation, which employ transformer-based and domain-adaptive approaches, show promise in addressing the complexity of ocular diseases such as diabetic retinopathy. However, current algorithms face challenges in effectively accommodating domain-specific variations and limitations of training datasets, which fail to represent real-world conditions comprehensively. Manual inspection by specialists remains time-consuming despite technological progress in medical imaging, underscoring the pressing need for automated and robust segmentation techniques. Additionally, these methods have deficiencies in handling unlabeled target sets, requiring extra preprocessing steps and manual intervention, which hinders their scalability and practical application in clinical settings. This research introduces a novel framework that employs semi-supervised domain adaptation and contrastive pre-training to address these limitations. The proposed model effectively learns from target data by implementing a novel pseudo-labeling approach and feature-based knowledge distillation within a temporal convolutional network (TCN) and extracts robust, domain-independent features. This approach enhances cross-domain adaptation, significantly enhancing the model's versatility and performance in clinical settings. The semi-supervised domain adaptation component overcomes the challenges posed by domain shifts, while pseudo-labeling utilizes the data's inherent structure for enhanced learning, which is particularly beneficial when labeled data is scarce. Evaluated on the DRIVE and CHASE_DB1 datasets, which contain clinical fundus images, the proposed model achieves outstanding performance, with accuracy, sensitivity, specificity, and AUC values of 0.9792, 0.8640, 0.9901, and 0.9868 on DRIVE, and 0.9830, 0.9058, 0.9888, and 0.9950 on CHASE_DB1, respectively, outperforming current state-of-the-art vessel segmentation methods. The partitioning of datasets into training and testing sets ensures thorough validation, while extensive ablation studies with thorough sensitivity analysis of the model's parameters and different percentages of labeled data further validate its robustness.

Versatile Teacher: A class-aware teacher–student framework for cross-domain adaptation

Addressing the challenge of domain shift between datasets is vital in maintaining model performance. In the context of cross-domain object detection, the teacher–student framework, a widely-used semi-supervised model, has shown significant accuracy improvements. However, existing methods often overlook class differences, treating all classes equally, resulting in suboptimal results. Furthermore, the integration of instance-level alignment with a one-stage detector, essential due to the absence of a Region Proposal Network (RPN), remains unexplored in this framework. In response to these shortcomings, we introduce a novel teacher–student model named Versatile Teacher (VT). VT differs from previous works by considering class-specific detection difficulty and employing a two-step pseudo-label selection mechanism, referred to as Class-aware Pseudo-label Adaptive Selection (CAPS), to generate more reliable pseudo labels. These labels are leveraged as saliency matrices to guide the discriminator for targeted instance-level alignment. Our method demonstrates promising results on three benchmark datasets, and extends the alignment methods for widely-used one-stage detectors, presenting significant potential for practical applications. Code is available at https://github.com/RicardooYoung/VersatileTeacher.

Marginalized Augmented Few-Shot Domain Adaptation.

Domain adaptation (DA) has recently drawn a lot of attention, as it facilitates unlabeled target learning by borrowing knowledge from an external source domain. Most existing DA solutions seek to align feature representations between the labeled source and unlabeled target data. However, the scarcity of target data easily results in negative transfer, as it misleads the cross DA to the dominance of the source. To address the challenging few-shot domain adaptation (FSDA) problem, in this article, we propose a novel marginalized augmented FSDA (MAF) approach to address the cross-domain distribution disparity and insufficiency of target data simultaneously. On the one hand, cross-domain continuity augmentation (CCA) synthesizes abundant intermediate patterns across domains leading to a continuous domain-invariant latent space. On the other hand, sufficient source-supervised semantic augmentation (SSA) is explored to progressively diversify the conditional distribution within and across domains. Moreover, the proposed augmentation strategies are implemented efficiently via an expected transferable cross-entropy (CE) loss over the augmented distribution instead of explicit data synthesis, and minimizing the upper bound of the expected loss introduces negligible extra computing cost. Experimentally, our method outperforms the state of the art in various FSDA benchmarks, which demonstrates the effectiveness and contribution of our work. Our source code is provided at https://github.com/scottjingtt/MAF.git.

Cross-Domain Adaptation Techniques for Robust Plant Disease Detection: A DANN-CORAL Hybrid Approach

Plant disease detection with deep learning models has shown promising results, but these models often struggle with generalizing across diverse agricultural environments due to domain shifts in imaging conditions. This paper presents a novel hybrid approach focusing on cross-domain adaptation techniques to address the challenge of domain shift. Our proposed method combines the Domain-Adversarial Neural Network (DANN) with Correlation Alignment (CORAL) to mitigate domain shifts between datasets. The DANN framework enforces domain-invariant feature learning through adversarial training. Using the PlantVillage Dataset, with controlled environment images, and the New Plant Village Dataset, with varied conditions, the model is first trained on PlantVillage and then adapted to New Plant Village using the CORAL loss to support the second-order statistics. In case of domain shift experiments with various datasets, DANN-CORAL achieved accuracies 91.39%, precision 93.36%, recall 88.9% and F1-scores 91.05% indicating the robustness and generalizability of our model is better than the other baseline models. This approach enhances model robustness and adaptability, providing insights into combining adversarial and statistical alignment for cross-domain adaptation in agricultural imaging.

An Unsupervised Domain-Adaptive Framework for Chinese Spelling Checking

Chinese Spelling Check (CSC) is a meaningful task in the area of Natural Language Processing (NLP), which aims at detecting spelling errors in Chinese texts and then correcting these errors. Current typical Chinese Spelling Check models have shown impressive performance in general datasets with the help of pretrained language models such as BERT, but suffer great perform loss in downstream tasks with domain-specific terms because they are primarily trained on general corpora. To verify the cross-domain adaptation ability of these models, we build three new datasets with abundant domain-specific terms on financial, medical, and legal domains and conduct empirical investigations on them in the corresponding domain-specific test datasets to verify the cross-domain adaptation ability. In response to the poor performance of the existing models, we propose a framework named uChecker which utilizes unsupervised method in spelling error detection and correction. Experiment results prove that uChecker can perform well in domain-specific test datasets while not losing its performance in the general domain.

Transfer learning with cross-domain adaptation of vibro-acoustic signals for electric motor mechanical fault detection and diagnosis.

Contact-based vibration measurement techniques involving accelerometers are typically utilized for machine health monitoring and manufacturing end-of-line quality control. Acoustic signals, encompassing sound pressure and particle velocity, carry crucial information about the operational status of underlying equipment and can be utilized for detecting various machine faults. Utilizing one-dimensional Convolution Neural Networks (1D-CNN) based on deep learning for processing raw acoustic time signals has proven to be a valuable approach in detecting machinery faults, offering an effective alternative to using traditional accelerometer signals. However, the success of deep learning methods is heavily dependent on the quantity and availability of data for robust network training. When transitioning to new non-contact type sensor technology, manufacturers may face challenges due to insufficient data for training deep learning models. In such cases, leveraging existing historical accelerometer-based data becomes crucial. Transfer learning emerges as a promising solution, allowing deep learning models trained on a source domain (SD) to be applied to a separate but related target domain (TD). This paper hence introduces a domain adaptation methodology by implementing transfer learning, where the SD 1D-CNN network is trained on accelerometer signals, then the trained model weights are transferred to the TD 1D-CNN network that can process acoustic signals.

Active Bidirectional Self-Training Network for Cross-Domain Segmentation in Remote-Sensing Images

Semantic segmentation with cross-domain adaptation in remote-sensing images (RSIs) is crucial and mitigates the expense of manually labeling target data. However, the performance of existing unsupervised domain adaptation (UDA) methods is still significantly impacted by domain bias, leading to a considerable gap compared to supervised trained models. To address this, our work focuses on semi-supervised domain adaptation, selecting a small subset of target annotations through active learning (AL) that maximize information to improve domain adaptation. Overall, we propose a novel active bidirectional self-training network (ABSNet) for cross-domain semantic segmentation in RSIs. ABSNet consists of two sub-stages: a multi-prototype active region selection (MARS) stage and a source-weighted class-balanced self-training (SCBS) stage. The MARS approach captures the diversity in labeled source data by introducing multi-prototype density estimation based on Gaussian mixture models. We then measure inter-domain similarity to select complementary and representative target samples. Through fine-tuning with the selected active samples, we propose an enhanced self-training strategy SCBS, designed for weighted training on source data, aiming to avoid the negative effects of interfering samples. We conduct extensive experiments on the LoveDA and ISPRS datasets to validate the superiority of our method over existing state-of-the-art domain-adaptive semantic segmentation methods.

Cross-domain continual learning via CLAMP

Artificial neural networks, celebrated for their human-like cognitive learning abilities, often encounter the well-known catastrophic forgetting (CF) problem, where the neural networks lose the proficiency in previously acquired knowledge. Despite numerous efforts to mitigate CF, it remains the significant challenge particularly in complex changing environments. This challenge is even more pronounced in cross-domain adaptation following the continual learning (CL) setting, which is a more challenging and realistic scenario that is under-explored. To this end, this article proposes a cross-domain CL approach making possible to deploy a single model in such environments without additional labelling costs. Our approach, namely continual learning approach for many processes (CLAMP), integrates a class-aware adversarial domain adaptation strategy to align a source domain and a target domain. An assessor-guided learning process is put forward to navigate the learning process of a base model assigning a set of weights to every sample controlling the influence of every sample and the interactions of each loss function in such a way to balance the stability and plasticity dilemma thus preventing the CF problem. The first assessor focuses on the negative transfer problem rejecting irrelevant samples of the source domain while the second assessor prevents noisy pseudo labels of the target domain. Both assessors are trained in the meta-learning approach using random transformation techniques and similar samples of the source domain. Theoretical analysis and extensive numerical validations demonstrate that CLAMP significantly outperforms established baseline algorithms across all experiments by at least 10% margin.

Optimal heterogeneous domain adaptation for text classification in transfer learning

The prime challenge of unsupervised symmetric heterogeneous cross-domain adaptation is to train the source domain and apply the trained knowledge to the target domain. Most of the existing algorithms for unsupervised transfer learning create the subspace of the source domain features and target domain features for training purposes. It is an extensive computational process as most of the techniques require labeled source data. Many techniques also suffer from the loss of originality of features in both domains. This paper aims to consider the feature vectors of both the source and target domain for training the data based on the similarity of exemplar (feature) vectors of different instances, known as Instance Similarity Feature (ISF). The use of vectorization method for the similarity of features is proposed in this paper. The exemplar vectors are chosen randomly for the target datasets. Hence, to acquire relevant factual data in the knowledge base for training in our research, we worked to increase the domain separation error between source and target instances. To avoid the instability caused due to poor exemplar vector selection, the K-means clustering approach is followed after feature similarity, known as K-means Instance Similarity Feature (KISF). Many existing transfer learning techniques are based on the original feature set, which can cause degeneracy, hence affecting Accuracy. In order to vanquish the limitations of existing approaches, we have introduced novel optimal models with KISF and Ant Lion Optimizer (KISFA), KISF with Particle Swarm Optimization (KISFP) and KISF with Biogeography Based Optimization (KISFB). High-dimensionality can impact efficacy of the model, hence, feature selection with nature-based optimizer namely: Ant Lion Optimizer, Particle Swarm Optimization and Biogeography-Based Optimization are applied. We measure the performance of the proposed models by using Support Vector Machine, Logistic Regression, Random Forest, Naive Baye’s, K-Nearest Neighbor and Decision Tree as classifiers, and Accuracy and F1-score as fitness functions. Extensive experiments are performed on four datasets with 50 iterations. The proposed model is compared with eleven other techniques and our technique outperforms all other techniques in average Accuracy. The validation is performed on the dataset using 10-fold cross-validation. The statistical test was performed using ANOVA, proving that our technique is significantly better than other techniques.

Towards multi-scene learning: A novel cross-domain adaptation model based on sparse filter for traction motor bearing fault diagnosis in high-speed EMU

Fault diagnosis of traction motor bearing is of great significance to improve the reliability and safety of high-speed electric multiple units (EMU). While the fault diagnosis method based on cross-domain adaptation has been successful in scenarios involving speed or load fluctuations, existing methods ignore the independence and diversity of features, resulting in unsatisfactory diagnostic results under multi-scene learning, thereby reducing the generalization ability. Moreover, the development of complex models is time-consuming, and their computational efficiency is low. To address these issues, this study proposes a novel cross-domain adaptation model based on sparse filtering (SFCDA), which consists of only two fully connected (FC) layers. Firstly, pre-training is conducted to utilize the soft reconstruction penalties to constrain the weights of sparse filtering and improve the independence of features. The weights of unsupervised training are used to initialize the parameters of the first FC layer of the SFCDA model. Secondly, a multiple sparse regularization (MSR) algorithm is proposed and used to constrain the SFCDA. Then, fine-tuning is conducted, in which the structural alignment function is used to measure the distribution distance between the source and target domain data. Minimizing the kernel norm can improve the diversity of features and enhance the robustness. Finally, the effectiveness of SFCDA in multi-scene learning is proved theoretically. It is validated in three fault diagnosis scenes in four different bearing datasets. The results show that the suggested approach is more straightforward and has a better fault diagnosis effect than the state-of-the-art domain adaptive approaches.

CloudMix: Dual Mixup Consistency for Unpaired Point Cloud Completion.

Due to the unsatisfactory performance of supervised methods on unpaired real-world scans, point cloud completion via cross-domain adaptation has recently drawn growing attention. Nevertheless, previous approaches only focus on alleviating the distribution shift through domain alignment, resulting in massive information loss of real-world domain data. To tackle this issue, we propose a dual mixup-induced consistency regularization to integrate both source and target domain to improve robustness and generalization capability. Specifically, we mix up virtual and real-world shapes in the input and latent feature space respectively, and then regularize the completion network by forcing two kinds of mixed completion predictions to be consistent. To further adapt to each instance within the real-world domain, we design a novel density-aware refiner to utilize local context information to preserve the fine-grained details and remove noise or outliers for coarse completion. Extensive experiments on real-world scans and our synthetic unpaired datasets demonstrate the superiority of our method over existing state-of-the-art approaches.

RD-OpenMax: Rethinking OpenMax for Robust Realistic Open-Set Recognition.

Open-set recognition (OSR) toward a practical open-world setting has attracted increasing research attention in recent years. However, existing OSR settings are either too idealized or focus on specific scenes such as long-tailed distribution and few-shot samples, which fail to capture the complexity of real-world scenarios. In this article, we propose a realistic OSR (ROSR) setting that covers a diverse range of challenging and real-world scenarios, including fine-grained cases with strong semantic correlation and a large number of species, few-shot samples, long-tailed sample distribution, dynamic inputs (e.g., images, spatio-temporal, and multimodal signals) and cross-domain adaptation. In particular, we rethink the simple and basic OpenMax for the ROSR setting and introduce a novel method, regularized discriminative OpenMax (RD-OpenMax), to handle the challenges in the ROSR setting. RD-OpenMax improves upon the basic OpenMax approach by introducing a covariance attention-based covariance pooling (CACP) module as a global aggregation step before the deep architecture's classifier. This module explores rich statistical information on features and provides discriminative distance scores for OpenMax. To address the instability of extreme value theory (EVT) estimation due to insufficient training samples under few-shot and long-tailed scenarios, we propose a regularized EVT (REVT) method based on Monte Carlo sampling to recalibrate the distribution of distance scores. As such, our RD-OpenMax performs a REVT model of distance scores generated by discriminative CACP representations to distinguish known classes and recognize unknown ones effectively and robustly. Extensive experiments are conducted on more than ten visual benchmarks across several scenarios, and the empirical comparisons show that the ROSR setting challenges existing state-of-the-art OSR approaches. Moreover, our RD-OpenMax clearly outperforms its counterparts under the ROSR setting while performing favorably against state-of-the-arts under the traditional OSR setting.

REACT: Remainder Adaptive Compensation for Domain Adaptive Object Detection.

Domain adaptive object detection (DAOD) aims to infer a robust detector on the target domain with the labelled source datasets. Recent studies utilize a feature extractor shared on the source and target domains to capture the domain-invariant features and the task-relevant information with both feature-alignment constraint and source annotations. However, the feature extractor shared across domains discards partial task-relevant information of the target domain due to the domain gap and lack of target annotations, leading to compromised discrimination capabilities within target domain. To this end, we propose a novel REmainder Adaptive CompensaTion network (REACT) to adaptively compensate the extracted features with the remainder features for generating task-relevant features. The key insight is that the remainder features contain the discarded task-relevant information, so they can be adapted to compensate for the inadequate target features. Especially, REACT introduces an additional remainder branch to regain the remainder features, and then adaptively utilizes them to compensate for the discarded task-relevant information, improving discrimination on the target domain. Extensive experiments over multiple cross-domain adaptation tasks with three baselines demonstrate that our approach gains significant improvements and achieves superior performance compared with highly-optimized state-of-the-art methods.

A novel cross-domain adaptation framework for unsupervised criminal jargon detection via pre-trained contextual embedding of darknet corpus

As the regulation on the surface web becomes more stringent, criminals are gradually turning to the darknet markets for illicit operations. Moderating and studying the content on the marketplaces contribute to the combat of criminal forces in the darknet. Nevertheless, to evade the surveillance of law enforcement, jargons are widely used in criminal conversations as a disguise. These jargons misinterpret the meaning of seemingly innocuous words in cryptic ways, creating a huge challenge for criminal investigation. Current research on Chinese jargon detection focuses on keyword matching. However, this approach cannot keep up with the rapid update of new jargons from various domains. To the best of our knowledge, we are the first to conduct Chinese jargons detection research in the darknet markets. Specifically, we design an unsupervised cross-domain adaptation Chinese jargon detection framework (CJD-Framework) integrated with the pre-trained language model. Firstly, six underground markets in Chinese are crawled to build the first dataset of darknet corpus (DC-dataset). Next, a pre-training model based on Chinese word is proposed to extract contextual embeddings for darknet words. Finally, relying on semantic similarity analysis, a cross-corpus framework is constructed to effectively identify Chinese jargons in the darknet. Comprehensive experiments demonstrate the effectiveness and generalizability of the CJD-framework over the state-of-the-art models, with a detection accuracy of 91.5%. The darknet corpus dataset and innovative framework proposed in this research can provide sources and ideas for future analysis of underground crimes in the darknet markets.

A Systematic Review of Deep Learning Approaches for Vessel Segmentation in Retinal Fundus Images

Retinal vessel segmentation (RVS) is important to accurately differentiate retinal vasculature for diagnosing and monitoring various ocular and systemic diseases. The traditional methods for RVS have mostly involved supervised learning, although semi-supervised and unsupervised techniques are on the rise. This paper reviews the increase in complexity of developments in RVS primarily after 2020. The methods were chosen to cover both the gradual transition over time and a variety of unorthodox or combinatorial approaches. This includes convolutional neural networks, encoder-decoder models, generative models, and other multi-modal or hybrid techniques. CNN approaches discussed employ Zero Phase Component Analysis, Global Contrast Normalization, and reinforcement learning. Encoder-decoder models include approaches such as the use of skip and residual connections, spatial attention, and atrous enhancement U-Net. Generative models propose short link connections, recurrent residual blocks, and multi-scale features to refine convolutional blocks. Hybrid methods involve the use of connectivity features, the MISODATA Algorithm, cross-domain adaptation, and multiple filters (such as morphological, match, and Gabor). All the frameworks are compared based on their performance on the benchmark dataset DRIVE to provide a comprehensive understanding of the current state of RVS.

Autonomous Cross Domain Adaptation Under Extreme Label Scarcity.

A cross domain multistream classification is a challenging problem calling for fast domain adaptations to handle different but related streams in never-ending and rapidly changing environments. Notwithstanding that existing multistream classifiers assume no labeled samples in the target stream, they still incur expensive labeling costs since they require fully labeled samples of the source stream. This article aims to attack the problem of extreme label shortage in the cross domain multistream classification problems where only very few labeled samples of the source stream are provided before process runs. Our solution, namely, Learning Streaming Process from Partial Ground Truth (LEOPARD), is built upon a flexible deep clustering network where its hidden nodes, layers, and clusters are added and removed dynamically with respect to varying data distributions. A deep clustering strategy is underpinned by a simultaneous feature learning and clustering technique leading to clustering-friendly latent spaces. A domain adaptation strategy relies on the adversarial domain adaptation technique where a feature extractor is trained to fool a domain classifier by classifying source and target streams. Our numerical study demonstrates the efficacy of LEOPARD where it delivers improved performances compared to prominent algorithms in 15 of 24 cases. Source codes of LEOPARD are shared in https://github.com/wengweng001/LEOPARD.git to enable further study.

An application of stereo matching algorithm based on transfer learning on robots in multiple scenes

Robot vision technology based on binocular vision holds tremendous potential for development in various fields, including 3D scene reconstruction, target detection, and autonomous driving. However, current binocular vision methods used in robotics engineering have limitations such as high costs, complex algorithms, and low reliability of the generated disparity map in different scenes. To overcome these challenges, a cross-domain stereo matching algorithm for binocular vision based on transfer learning was proposed in this paper, named Cross-Domain Adaptation and Transfer Learning Network (Ct-Net), which has shown valuable results in multiple robot scenes. First, this paper introduces a General Feature Extractor to extract rich general feature information for domain adaptive stereo matching tasks. Then, a feature adapter is used to adapt the general features to the stereo matching network. Furthermore, a Domain Adaptive Cost Optimization Module is designed to optimize the matching cost. A disparity score prediction module was also embedded to adaptively adjust the search range of disparity and optimize the cost distribution. The overall framework was trained using a phased strategy, and ablation experiments were conducted to verify the effectiveness of the training strategy. Compared with the prototype PSMNet, on KITTI 2015 benchmark, the 3PE-fg of Ct-Net in all regions and non-occluded regions decreased by 19.3 and 21.1% respectively, meanwhile, on the Middlebury dataset, the proposed algorithm improves the sample error rate at least 28.4%, which is the Staircase sample. The quantitative and qualitative results obtained from Middlebury, Apollo, and other datasets demonstrate that Ct-Net significantly improves the cross-domain performance of stereo matching. Stereo matching experiments in real-world scenes have shown that it can effectively address visual tasks in multiple scenes.

Open-Ended Diverse Solution Discovery with Regulated Behavior Patterns for Cross-Domain Adaptation

Open-Ended Diverse Solution Discovery with Regulated Behavior Patterns for Cross-Domain Adaptation

Unsupervised domain adaptation for semantic segmentation via cross-region alignment

Semantic segmentation requires a lot of training data, which necessitates costly annotation. There have been many studies on unsupervised domain adaptation (UDA) from one domain to another, e.g., from computer graphics to real images. However, there is still a gap in accuracy between UDA and supervised training on native domain data. It is arguably attributable to the class-level misalignment between the source and target domain data. To cope with this, we propose a method that applies adversarial training to align two feature distributions in the target domain. It uses a self-training framework to split the image into two regions (i.e., trusted and untrusted), which form two distributions to align in the feature space. We term this approach cross-region adaptation (CRA) to distinguish it from the previous methods of aligning different domain distributions, which we call cross-domain adaptation (CDA). CRA can be applied after any CDA method. Experimental results show that this always improves the accuracy of the combined CDA method.

Transferable adversarial masked self-distillation for unsupervised domain adaptation

Unsupervised domain adaptation (UDA) aims to transfer knowledge from a labeled source domain to a related unlabeled target domain. Most existing works focus on minimizing the domain discrepancy to learn global domain-invariant representation using CNN-based architecture while ignoring both transferable and discriminative local representation, e.g, pixel-level and patch-level representation. In this paper, we propose the Transferable Adversarial Masked Self-distillation based on Vision Transformer architecture to enhance the transferability of UDA, named TAMS. Specifically, TAMS jointly optimizes three objectives to learn both task-specific class-level global representation and domain-specific local representation. First, we introduce adversarial masked self-distillation objective to distill representation from a full image to the representation predicted from a masked image, which aims to learn task-specific global class-level representation. Second, we introduce masked image modeling objectives to learn local pixel-level representation. Third, we introduce an adversarial weighted cross-domain adaptation objective to capture discriminative potentials of patch tokens, which aims to learn both transferable and discriminative domain-specific patch-level representation. Extensive studies on four benchmarks and the experimental results show that our proposed method can achieve remarkable improvements compared to previous state-of-the-art UDA methods.