Cross-domain Setting Research Articles

An Authenticated Controlled Quantum Dialogue Protocol Using Double-Linked GHZ-Like States in Cross-domain Setting

An Authenticated Controlled Quantum Dialogue Protocol Using Double-Linked GHZ-Like States in Cross-domain Setting

Are LSTMs good few-shot learners?

Deep learning requires large amounts of data to learn new tasks well, limiting its applicability to domains where such data is available. Meta-learning overcomes this limitation by learning how to learn. Hochreiter et al. (International conference on artificial neural networks, Springer, 2001) showed that an LSTM trained with backpropagation across different tasks is capable of meta-learning. Despite promising results of this approach on small problems, and more recently, also on reinforcement learning problems, the approach has received little attention in the supervised few-shot learning setting. We revisit this approach and test it on modern few-shot learning benchmarks. We find that LSTM, surprisingly, outperform the popular meta-learning technique MAML on a simple few-shot sine wave regression benchmark, but that LSTM, expectedly, fall short on more complex few-shot image classification benchmarks. We identify two potential causes and propose a new method called Outer Product LSTM (OP-LSTM) that resolves these issues and displays substantial performance gains over the plain LSTM. Compared to popular meta-learning baselines, OP-LSTM yields competitive performance on within-domain few-shot image classification, and performs better in cross-domain settings by 0.5–1.9% in accuracy score. While these results alone do not set a new state-of-the-art, the advances of OP-LSTM are orthogonal to other advances in the field of meta-learning, yield new insights in how LSTM work in image classification, allowing for a whole range of new research directions. For reproducibility purposes, we publish all our research code publicly.

DIWS-LCR-Rot-hop++: A Domain-Independent Word Selector for Cross-Domain Aspect-Based Sentiment Classification

The Aspect-Based Sentiment Classification (ABSC) models often suffer from a lack of training data in some domains. To exploit the abundant data from another domain, this work extends the original state-of-the-art LCR-Rot-hop++ model that uses a neural network with a rotatory attention mechanism for a cross-domain setting. More specifically, we propose a Domain-Independent Word Selector (DIWS) model that is used in combination with the LCR-Rot-hop++ model (DIWS-LCR-Rot-hop++). DIWS-LCR-Rot-hop++ uses attention weights from the domain classification task to determine whether a word is domain-specific or domain-independent, and discards domain-specific words when training and testing the LCR-Rot-hop++ model for cross-domain ABSC. Overall, our results confirm that DIWS-LCR-Rot-hop++ outperforms the original LCR-Rot-hop++ model under a cross-domain setting in case we impose an optimal domain-dependent attention threshold value for deciding whether a word is domain-specific or domain-independent. For a target domain that is highly similar to the source domain, we find that imposing moderate restrictions on classifying domain-independent words yields the best performance. Differently, a dissimilar target domain requires a strict restriction that classifies a small proportion of words as domain-independent. Also, we observe information loss which deteriorates the performance of DIWS-LCR-Rot-hop++ when we categorize an excessive amount of words as domain-specific and discard them.

Robust Prototypical Few-Shot Organ Segmentation with Regularized Neural-ODEs.

Despite the tremendous progress made by deep learning models in image semantic segmentation, they typically require large annotated examples, and increasing attention is being diverted to problem settings like Few-Shot Learning (FSL) where only a small amount of annotation is needed for generalisation to novel classes. This is especially seen in medical domains where dense pixel-level annotations are expensive to obtain. In this paper, we propose Regularized Prototypical Neural Ordinary Differential Equation (R-PNODE), a method that leverages intrinsic properties of Neural-ODEs, assisted and enhanced by additional cluster and consistency losses to perform Few-Shot Segmentation (FSS) of organs. R-PNODE constrains support and query features from the same classes to lie closer in the representation space thereby improving the performance over the existing Convolutional Neural Network (CNN) based FSS methods. We further demonstrate that while many existing Deep CNN-based methods tend to be extremely vulnerable to adversarial attacks, R-PNODE exhibits increased adversarial robustness for a wide array of these attacks. We experiment with three publicly available multi-organ segmentation datasets in both in-domain and cross-domain FSS settings to demonstrate the efficacy of our method. In addition, we perform experiments with seven commonly used adversarial attacks in various settings to demonstrate R-PNODE's robustness. R-PNODE outperforms the baselines for FSS by significant margins and also shows superior performance for a wide array of attacks varying in intensity and design.

Disparity refinement framework for learning-based stereo matching methods in cross-domain setting for laparoscopic images.

Stereo matching methods that enable depth estimation are crucial for visualization enhancement applications in computer-assisted surgery. Learning-based stereo matching methods have shown great promise in making accurate predictions on laparoscopic images. However, they require a large amount of training data, and their performance may be degraded due to domain shifts. Maintaining robustness and improving the accuracy of learning-based methods are still open problems. To overcome the limitations of learning-based methods, we propose a disparity refinement framework consisting of a local disparity refinement method and a global disparity refinement method to improve the results of learning-based stereo matching methods in a cross-domain setting. Those learning-based stereo matching methods are pre-trained on a large public dataset of natural images and are tested on two datasets of laparoscopic images. Qualitative and quantitative results suggest that our proposed disparity framework can effectively refine disparity maps when they are noise-corrupted on an unseen dataset, without compromising prediction accuracy when the network can generalize well on an unseen dataset. Our proposed disparity refinement framework could work with learning-based methods to achieve robust and accurate disparity prediction. Yet, as a large laparoscopic dataset for training learning-based methods does not exist and the generalization ability of networks remains to be improved, the incorporation of the proposed disparity refinement framework into existing networks will contribute to improving their overall accuracy and robustness associated with depth estimation.

Federated Learning on Non-IID Graphs via Structural Knowledge Sharing

Graph neural networks (GNNs) have shown their superiority in modeling graph data. Owing to the advantages of federated learning, federated graph learning (FGL) enables clients to train strong GNN models in a distributed manner without sharing their private data. A core challenge in federated systems is the non-IID problem, which also widely exists in real-world graph data. For example, local data of clients may come from diverse datasets or even domains, e.g., social networks and molecules, increasing the difficulty for FGL methods to capture commonly shared knowledge and learn a generalized encoder. From real-world graph datasets, we observe that some structural properties are shared by various domains, presenting great potential for sharing structural knowledge in FGL. Inspired by this, we propose FedStar, an FGL framework that extracts and shares the common underlying structure information for inter-graph federated learning tasks. To explicitly extract the structure information rather than encoding them along with the node features, we define structure embeddings and encode them with an independent structure encoder. Then, the structure encoder is shared across clients while the feature-based knowledge is learned in a personalized way, making FedStar capable of capturing more structure-based domain-invariant information and avoiding feature misalignment issues. We perform extensive experiments over both cross-dataset and cross-domain non-IID FGL settings, demonstrating the superiority of FedStar.

Attention-Enabled Ensemble Deep Learning Models and Their Validation for Depression Detection: A Domain Adoption Paradigm

Depression is increasingly prevalent, leading to higher suicide risk. Depression detection and sentimental analysis of text inputs in cross-domain frameworks are challenging. Solo deep learning (SDL) and ensemble deep learning (EDL) models are not robust enough. Recently, attention mechanisms have been introduced in SDL. We hypothesize that attention-enabled EDL (aeEDL) architectures are superior compared to attention-not-enabled SDL (aneSDL) or aeSDL models. We designed EDL-based architectures with attention blocks to build eleven kinds of SDL model and five kinds of EDL model on four domain-specific datasets. We scientifically validated our models by comparing “seen” and “unseen” paradigms (SUP). We benchmarked our results against the SemEval (2016) sentimental dataset and established reliability tests. The mean increase in accuracy for EDL over their corresponding SDL components was 4.49%. Regarding the effect of attention block, the increase in the mean accuracy (AUC) of aeSDL over aneSDL was 2.58% (1.73%), and the increase in the mean accuracy (AUC) of aeEDL over aneEDL was 2.76% (2.80%). When comparing EDL vs. SDL for non-attention and attention, the mean aneEDL was greater than aneSDL by 4.82% (3.71%), and the mean aeEDL was greater than aeSDL by 5.06% (4.81%). For the benchmarking dataset (SemEval), the best-performing aeEDL model (ALBERT+BERT-BiLSTM) was superior to the best aeSDL (BERT-BiLSTM) model by 3.86%. Our scientific validation and robust design showed a difference of only 2.7% in SUP, thereby meeting the regulatory constraints. We validated all our hypotheses and further demonstrated that aeEDL is a very effective and generalized method for detecting symptoms of depression in cross-domain settings.

Task context transformer and GCN for few-shot learning of cross-domain

Cross-Domain Few-Shot Learning (CD-FSL) to recognize new categories in a new domain with few samples has attracted significant attention. Recently, task-specific CD-FSL emerges as promising research for its great generalization. However, the existing task-specific CD-FSL methods are not robust enough in task context modeling and task-specific feature learning, especially under the cross-domain setting. To tackle this problem, a Task Context Transformer and Graph Convolutional Network (TCT-GCN) method for CD-FSL is proposed. The proposed TCT-GCN constructs three modules: 1) Multi-Level Feature Fusion, which fuses domain-shared low-level features into domain-unshared high-level features; 2) Transformer-based Task Context Encoder, which models sample order-independent task context; 3) Graph Convolutional Network-based Adaptive Feature Learner, which adaptively learns task-specific features. Experiments on eight CD-FSL datasets reveal the effectiveness of our TCT-GCN.

A Disparity Refinement Framework for Learning-based Stereo Matching Methods in Cross-domain Setting for Laparoscopic Images.

Stereo matching methods that enable depth estimation are crucial for visualization enhancement applications in computer-assisted surgery (CAS). Learning-based stereo matching methods are promising to predict accurate results for applications involving video images. However, they require a large amount of training data, and their performance may be degraded due to domain shifts. Maintaining robustness and improving performance of learning-based methods are still open problems. To overcome the limitations of learning-based methods, we propose a disparity refinement framework consisting of a local disparity refinement method and a global disparity refinement method to improve the results of learning-based stereo matching methods in a cross-domain setting. Those learning-based stereo matching methods are pre-trained on a large public dataset of natural images and are tested on a dataset of laparoscopic images. Results from the SERV-CT dataset showed that our proposed framework can effectively refine disparity maps on an unseen dataset even when they are corrupted by noise, and without compromising correct prediction, provided the network can generalize well on unseen datasets. As such, our proposed disparity refinement framework has the potential to work with learning-based methods to achieve robust and accurate disparity prediction. Yet, as a large laparoscopic dataset for training learning-based methods does not exist and the generalization ability of networks remains to be improved, it will be beneficial to incorporate the proposed disparity refinement framework into existing networks for more accurate and robust depth estimation.

Parallel Split-Join Networks for Shared Account Cross-Domain Sequential Recommendations

Sequential recommendation is a task in which one models and uses sequential information about user behavior for recommendation purposes. We study sequential recommendation in a context in which multiple individual users share a single account (i.e., they have a shared account) and in which user behavior is available in multiple domains (i.e., recommendations are cross-domain). These two characteristics bring new challenges on top of those of the traditional sequential recommendation task. First, we need to identify the behavior associated with different users and different user roles under the same account in order to recommend the right item to the right user role at the right time. Second, we need to identify behavior in one domain that might be helpful to improve recommendations in other domains. We study <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">shared account cross-domain sequential recommendation</i> and propose a <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">p</b> arallel <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">s</b> plit- <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">j</b> oin <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Net</b> work (Parallel Split-Join Network (PSJNet)), a parallel modeling network to address the two challenges above. We use “split” to address the challenge raised by shared accounts; PSJNet learns role-specific representations and uses a gating mechanism to filter out, from mixed user behavior, information of user roles that might be useful for another domain. In addition, “join” is used to address the challenge raised by the cross-domain setting; PSJNet learns cross-domain representations by combining the information from “split” and then transforms it to another domain. We present two variants of PSJNet: PSJNet-I and PSJNet-II. PSJNet-I is a “split-by-join” framework that splits the mixed representations to get role-specific representations and joins them to obtain cross-domain representations at each timestamp simultaneously. PSJNet-II is a “split-and-join” framework that first splits role-specific representations at each timestamp, and then the representations from all timestamps and all roles are joined to obtain cross-domain representations. We concatenate the in-domain and cross-domain representations to compute a recommendation score for each item. Both PSJNet-I and PSJNet-II can simultaneously generate recommendations for two domains where user behavior in two domains is synchronously shared at each timestamp. We use two datasets to assess the effectiveness of PSJNet. The first dataset is a simulated shared account cross-domain sequential recommendation dataset obtained by randomly merging the Amazon logs from different users in the movie and book domains. The second dataset is a real-world shared account cross-domain sequential recommendation dataset built from smart TV watching logs of a commercial organization. Our experimental results demonstrate that PSJNet outperforms state-of-the-art sequential recommendation baselines in terms of MRR and Recall.

CD-BCM:Cross-Domain Batch Certificates Management Based On Blockchain

AbstractWith the development of information networks, the entities from different network domains interact with each other more and more frequently. Therefore, identity management and authentication are essential in cross-domain setting. The traditional Public Key Infrastructure (PKI) architecture has some problems, including single point of failure, inefficient certificate revocation status management and also lack of privacy protection, which cannot meet the demand of cross-domain identity authentication. Blockchain is suitable for multi-participant collaboration in multi-trust domain scenarios. In this paper, a cross-domain certificate management scheme CD-BCM based on the consortium blockchain is proposed. For the issue of Certificate Authority’s single point of failure, we design a multi-signature algorithm. In addition, we propose a unified structure for batch certificates verification and conversion, which improve the efficiency of erroneous certificate identification. Finally, by comparing with current related schemes, our scheme achieves good functionality and scalability in the scenario of cross-domain certificate management.

Unsupervised domain adaptation for the detection of cardiomegaly in cross-domain chest X-ray images.

In recent years, several deep learning approaches have been successfully applied in the field of medical image analysis. More specifically, different deep neural network architectures have been proposed and assessed for the detection of various pathologies based on chest X-ray images. While the performed assessments have shown very promising results, most of them consist in training and evaluating the performance of the proposed approaches on a single data set. However, the generalization of such models is quite limited in a cross-domain setting, since a significant performance degradation can be observed when these models are evaluated on data sets stemming from different medical centers or recorded under different protocols. The performance degradation is mostly caused by the domain shift between the training set and the evaluation set. To alleviate this problem, different unsupervised domain adaptation approaches are proposed and evaluated in the current work, for the detection of cardiomegaly based on chest X-ray images, in a cross-domain setting. The proposed approaches generate domain invariant feature representations by adapting the parameters of a model optimized on a large set of labeled samples, to a set of unlabeled images stemming from a different data set. The performed evaluation points to the effectiveness of the proposed approaches, since the adapted models outperform optimized models which are directly applied to the evaluation sets without any form of domain adaptation.

Self-taught cross-domain few-shot learning with weakly supervised object localization and task-decomposition

The shift between the source and target domains is the main challenge in cross-domain few-shot learning (CD-FSL) tasks. However, the target domain is unknown during training in the source domain, which results in a lack of directed guidance for target tasks. We observe that, owing to similar backgrounds in the target domains, self-labelled samples can be applied as prior tasks to transfer knowledge onto target tasks. Accordingly, we propose a task-expansion-decomposition framework for CD-FSL called the self-taught (ST) approach, which alleviates the problem of non-target guidance by constructing task-oriented metric spaces. Specifically, weakly supervised object localization (WSOL) and self-supervised technologies are employed to enrich task-oriented samples by exchanging and rotating discriminative regions, which generates a more abundant task set. Then, these tasks are decomposed into several tasks to complete few-shot recognition and rotation classification tasks. Transferring the source knowledge onto the target tasks and focusing on discriminative regions is beneficial to the task completion. We conducted extensive experiments under a cross-domain setting, including eight target domains: CUB, Cars, Places, Plantae, CropDieases, EuroSAT, ISIC, and ChestX. The experimental results demonstrate that the proposed ST approach is applicable to various metric-based models and provides promising improvements to CD-FSL.

Interpretable bilinear attention network with domain adaptation improves drug–target prediction

Predicting drug–target interaction is key for drug discovery. Recent deep learning-based methods show promising performance, but two challenges remain: how to explicitly model and learn local interactions between drugs and targets for better prediction and interpretation and how to optimize generalization performance of predictions on novel drug–target pairs. Here, we present DrugBAN, a deep bilinear attention network (BAN) framework with domain adaptation to explicitly learn pairwise local interactions between drugs and targets, and adapt in response to out-of-distribution data. DrugBAN works on drug molecular graphs and target protein sequences to perform prediction, with conditional domain adversarial learning to align learned interaction representations across different distributions for better generalization on novel drug–target pairs. Experiments on three benchmark datasets under both in-domain and cross-domain settings show that DrugBAN achieves the best overall performance against five state-of-the-art baseline models. Moreover, visualizing the learned bilinear attention map provides interpretable insights from prediction results.

Domain adaptive object detection with model-agnostic knowledge transferring

The development of deep learning techniques has greatly benefited CNN-based object detectors, leading to unprecedented progress in recent years. However, the distribution variance between training and testing domains causes significant performance degradation. Labeling data for new scenarios is costly and time-consuming, so most existing domain adaptation methods perform feature alignment through adversarial training. While this can improve the accuracy of detectors in unlabeled target domains, the unconstrained domain alignment also negatively transfers the feature distribution, which compromises the recognition ability of the model. To address this problem, we propose the Knowledge Transfer Network (KTNet), which consists of object intrinsic knowledge mining and category relational knowledge constraint modules. Specifically, a binary classifier shared by the source and target domains is designed to extract common attribute knowledge of objects, which can align foreground and background features from different data domains adaptively. Then, we construct relational knowledge graphs to explicitly constrain the category correlations in the source, target, and cross-domain settings. These two modules guide the detector to learn object-related and domain-invariant representations, enabling the proposed KTNet to perform well in four commonly-used cross-domain scenarios. Furthermore, the ablation experiments show that our method is scalable to more complex backbone networks and different detection architectures.

Multi-Scale Feature Map Aggregation and Supervised Domain Adaptation of Fully Convolutional Networks for Urban Building Footprint Extraction

Automated building footprint extraction requires the Deep Learning (DL)-based semantic segmentation of high-resolution Earth observation images. Fully convolutional networks (FCNs) such as U-Net and ResUNET are widely used for such segmentation. The evolving FCNs suffer from the inadequate use of multi-scale feature maps in their backbone of convolutional neural networks (CNNs). Furthermore, the DL methods are not robust in cross-domain settings due to domain-shift problems. Two scale-robust novel networks, namely MSA-UNET and MSA-ResUNET, are developed in this study by aggregating the multi-scale feature maps in U-Net and ResUNET with partial concepts of the feature pyramid network (FPN). Furthermore, supervised domain adaptation is investigated to minimise the effects of domain-shift between the two datasets. The datasets include the benchmark WHU Building dataset and a developed dataset with 5× fewer samples, 4× lower spatial resolution and complex high-rise buildings and skyscrapers. The newly developed networks are compared to six state-of-the-art FCNs using five metrics: pixel accuracy, adjusted accuracy, F1 score, intersection over union (IoU), and the Matthews Correlation Coefficient (MCC). The proposed networks outperform the FCNs in the majority of the accuracy measures in both datasets. Compared to the larger dataset, the network trained on the smaller one shows significantly higher robustness in terms of adjusted accuracy (by 18%), F1 score (by 31%), IoU (by 27%), and MCC (by 29%) during the cross-domain validation of MSA-UNET. MSA-ResUNET shows similar improvements, concluding that the proposed networks when trained using domain adaptation increase the robustness and minimise the domain-shift between the datasets of different complexity.

Asymmetric Modality Translation for Face Presentation Attack Detection

Face presentation attack detection (PAD) is an essentialmeasure to protect face recognition systems from being spoofed by malicious users and has attracted great attention from both academia and industry. Although most of the existing methods can achieve desired performance to some extent, the generalization issue of face presentation attack detection under cross-domain settings (e.g., the setting of unseen attacks and varying illumination) remains to be solved. In this paper, we propose a novel framework based on asymmetric modality translation for face presentation attack detection in bi-modality scenarios. Under the framework, we establish connections between two modality images of genuine faces. Specifically, a novel modality fusion scheme is presented that the image of one modality is translated to the other one through an asymmetric modality translator, then fused with its corresponding paired image. The fusion result is fed as the input to a discriminator for inference. The training of the translator is supervised by an asymmetric modality translation loss. Besides, an illumination normalization module based on Pattern of Local Gravitational Force (PLGF) representation is used to reduce the impact of illumination variation. We conduct extensive experiments on three public datasets, which validate that our method is effective in detecting various types of attacks and achieves state-of-the-art performance under different evaluation protocols.

Deep Margin-Sensitive Representation Learning for Cross-Domain Facial Expression Recognition

Cross-domain Facial Expression Recognition (FER) aims to safely transfer the learned knowledge from labeled source data to unlabeled target data, which is challenging due to the subtle difference between various expressions and the large discrepancy between domains. Existing methods mainly focus on reducing the domain shift for transferable features but fail to learn discriminative representations for recognizing facial expression, which may result in negative transfer under cross-domain settings. To this end, we propose a novel Deep Margin-Sensitive Representation Learning (DMSRL) framework, which can extract multi-level discriminative features during sematic-aware domain adaptation. Specifically, we design a semantic metric learning module based on the category prior of source data and generated pseudo labels of target data, which can facilitate discriminative intra-domain representation learning and transferable inter-domain knowledge discovery by enlarging the category margin. Moreover, we develop a mutual information minimization module by simultaneously distilling the domain-invariant components and eliminating the domain-sensitive ones, which benefits discriminative transferable feature learning by generating accurate pseudo target labels. Furthermore, instead of only utilizing the global features, we formulate a multi-level feature extracting module to concurrently get the local ones, which contain detailed information to distinguish the small changes among different expressions. These modules are jointly utilized in our DMSRL in an end-to-end manner to ensure the positive transfer of source knowledge. Extensive experimental results on seven databases demonstrate that our DMSRL can achieve superior performance against state-of-the-art baselines.

Learning coronary artery calcium scoring in coronary CTA from non-contrast CT using unsupervised domain adaptation.

Deep learning methods have demonstrated the ability to perform accurate coronary artery calcium (CAC) scoring. However, these methods require large and representative training data hampering applicability to diverse CT scans showing the heart and the coronary arteries. Training methods that accurately score CAC in cross-domain settings remains challenging. To address this, we present an unsupervised domain adaptation method that learns to perform CAC scoring in coronary CT angiography (CCTA) from non-contrast CT (NCCT). To address the domain shift between NCCT (source) domain and CCTA (target) domain, feature distributions are aligned between two domains using adversarial learning. A CAC scoring convolutional neural network is divided into a feature generator that maps input images to features in the latent space and a classifier that estimates predictions from the extracted features. For adversarial learning, a discriminator is used to distinguish the features between source and target domains. Hence, the feature generator aims to extract features with aligned distributions to fool the discriminator. The network is trained with adversarial loss as the objective function and a classification loss on the source domain as a constraint for adversarial learning. In the experiments, three data sets were used. The network is trained with 1,687 labeled chest NCCT scans from the National Lung Screening Trial. Furthermore, 200 labeled cardiac NCCT scans and 200 unlabeled CCTA scans were used to train the generator and the discriminator for unsupervised domain adaptation. Finally, a data set containing 313 manually labeled CCTA scans was used for testing. Directly applying the CAC scoring network trained on NCCT to CCTA led to a sensitivity of 0.41 and an average false positive volume 140 mm3/scan. The proposed method improved the sensitivity to 0.80 and reduced average false positive volume of 20 mm3/scan. The results indicate that the unsupervised domain adaptation approach enables automatic CAC scoring in contrast enhanced CT while learning from a large and diverse set of CT scans without contrast. This may allow for better utilization of existing annotated data sets and extend the applicability of automatic CAC scoring to contrast-enhanced CT scans without the need for additional manual annotations. The code is publicly available at https://github.com/qurAI-amsterdam/CACscoringUsingDomainAdaptation.

Towards generalizable person re-identification with a bi-stream generative model

Generalizable person re-identification (re-ID) has attracted growing attention due to its powerful adaptation capability in the unseen data domain. However, existing solutions often neglect either crossing cameras (e.g., illumination and resolution differences) or pedestrian misalignments (e.g., viewpoint and pose discrepancies), which easily leads to poor generalization capability when adapted to the new domain. In this paper, we formulate these difficulties as: 1) Camera-Camera (CC) problem, which denotes the various human appearance changes caused by different cameras; 2) Camera-Person (CP) problem, which indicates the pedestrian misalignments caused by the same identity person under different camera viewpoints or changing pose. To solve the above issues, we propose a Bi-stream Generative Model (BGM) to learn the fine-grained representations fused with camera-invariant global feature and pedestrian-aligned local feature, which contains an encoding network and two stream decoding sub-network. Guided by original pedestrian images, one stream is employed to learn a camera-invariant global feature for the CC problem via filtering cross-camera interference factors. For the CP problem, another stream learns a pedestrian-aligned local feature for pedestrian alignment using information-complete densely semantically aligned part maps. Moreover, a part-weighted loss function is presented to reduce the influence of missing parts on pedestrian alignment. Extensive experiments demonstrate that our method outperforms the state-of-the-art methods on the large-scale generalizable re-ID benchmarks, involving domain generalization setting and cross-domain setting.