Visual tracking has seen remarkable advancements, largely driven by the availability of large-scale training datasets that have enabled the development of highly accurate and robust algorithms. While significant progress has been made in tracking general objects, research on more challenging scenarios, such as tracking camouflaged objects, remains limited. Camouflaged objects, which blend seamlessly with their surroundings or other objects, present unique challenges for detection and tracking in complex environments. In critical fields like military, security, agriculture, and marine monitoring, accurately tracking camouflaged objects is essential. To address this gap, we introduce the Camouflaged Object Tracking Dataset (COTD), a specialized benchmark designed specifically for evaluating camouflaged object tracking methods. The COTD dataset comprises 200 sequences and approximately 80,000 frames, each annotated with detailed bounding boxes. Our evaluation of 20 existing tracking algorithms reveals significant deficiencies in their performance with camouflaged objects. To address these issues, we propose a novel tracking framework, HIPTrack-MLS, which demonstrates promising results in improving tracking performance for camouflaged objects. COTD and code are avialable at https://github.com/openat25/HIPTrack-MLS.

ABSTRACT Recently, the majority of palmprint recognition studies have focused on feature extraction while neglecting security issues. Among the various attack types, spoofing attack poses a significant threat due to high success rates and minimal technical requirements. In this study, we explore the differences between real and fake palmprint images. Based on these differences, we propose the concept of ‘high‐frequency spoofing regions’ to capture key discriminative spoofing clues. Specifically, the high‐frequency spoofing regions adaptation ( HFSRA ) model is proposed to address palmprint anti‐spoofing. The HFSRA consists of two key modules: the texture analysis module (TAM) and the spoofing attention module (SAM). In particular, the TAM divides the input feature map into several patches and evaluates the texture distribution within each patch. Next, the SAM dynamically constructs an attention map by mapping the texture distribution to an attention weight matrix. This adaptive structure forces the model to focus on high‐frequency spoofing regions, which improves the model's ability to extract meaningful spoofing clues effectively. Furthermore, we establish three experimental protocols for evaluating the performance of palmprint anti‐spoofing models. These protocols provide a standardized evaluation framework for future studies. Extensive experiments conducted under these protocols demonstrate the effectiveness and competitiveness of HFSRA.

Since the introduction of domain adaptation to fault diagnosis, it has greatly improved the performance of cross-domain fault diagnosis. However, previous methods require constructing independent frameworks for specific domain adaptation scenarios in fault diagnosis, which reduces efficiency and increases consumption. The difficulty lies in the fact that the classes of source and target domains in universal domain adaptation scenarios are not uniform and agnostic, leading to the inability to identify unknown fault modes. To address this issue, we propose a novel Simulating Unseen Classes (SUC) method. It is a general framework that does not depend on a priori knowledge of the classes of target domains. More data with new classes are generated to simulate the unknown fault classes based on source domain at both data and feature levels. Class gap and domain gap between source and target domains are effectively reduced to extract domain invariant features. Adequate experiments are conducted on three benchmark datasets and the results demonstrate that our method can outperform other methods by a large margin. Note to Practitioners—Intelligent fault diagnosis plays an important role in automation system. In practice, the changes in equipment, operating conditions and the environment raise the difficulty of cross-domain fault diagnosis. This paper proposes a novel SUC method for universal domain adaptation in cross-domain fault diagnosis. It is a more general but difficult scenario where the prior of source and target domains is agnostic. More data with new classes is generated effectively based on source domain at both data and feature levels. Class gap and domain gap are reduced to improve the accuracy of fault diagnosis. Our method provides a universal framework that can be readily applied to different domain adaptation (DA) scenarios, including closed set DA, partial DA, open set DA, and universal DA. Experimental results demonstrate the superiority of our method. In the future, we will investigate more challenging issues in fault diagnosis, such as scenarios where the data in the target domain is unavailable during training.

Deep learning-based palmprint recognition methods take performance to the next level. However, most current methods rely on samples with clean labels. Noisy labels are difficult to avoid in practical applications and may affect the reliability of models, which poses a big challenge. In this paper, we propose a novel Multi-stage Noisy Label Selection and Correction (MNLSC) framework to address this issue. Three stages are proposed to improve the robustness of palmprint recognition. Clean simple samples are firstly selected based on self-supervised learning. A Fourier-based module is constructed to select clean hard samples. A pototype-based module is further introduced for selecting noisy labels from the remaining samples and correcting them. Finally, the model is trained by using clean and corrected labels to improve the performance. Experiments are conducted on several constrained and unconstrained palmprint databases. The results demonstrate the superiority of our method over other methods in dealing with different noise rates. Compared with the baseline method, the accuracy can be improved by up to 33.45% when there are 60% noisy labels.

Palmprint recognition offers a promising solution for convenient and private authentication. However, the scarcity of large-scale palmprint datasets constrains its development and application. Recent approaches have sought to mitigate this issue by synthesizing palmprints based on Bézier curves. Due to the lack of paired data between curves and palmprints, it is difficult to generate curve-driven palmprints with precise identity. To address this challenge, we propose a novel Pixel and Feature Identity Guidance (PFIG) framework to synthesize realistic palmprints, whose IDs are strictly governed by the Bézier curves. In order to establish ID mapping, an ID Injection (IDI) module is constructed to synthesize pseudo-paired data. Two cross-domain ID consistency losses at pixel and feature levels are further proposed to strictly preserve the semantic information of the input ID curves. Experimental results demonstrate that our ID-guided approach can synthesize more realistic palmprints with controllable identities. Based on only 80,000 synthesized palmprints for pre-training, the recognition accuracy can be improved by more than 18% in terms of TAR@1e-6. When trained exclusively on synthetic data, our method achieves superior performance to existing synthetic approaches. The source code is available at https://github.com/YuchenZou/PFIG-Palm.

Deep learning-based palmprint recognition methods have made good progress and obtained promising performance. However, most of them are mainly focused on continuously improving the recognition accuracy, while ignore the privacy preserving, which is also extremely significant. In this paper, we propose a novel Federated Metric Learning (FedML) method to address the issue of data privacy and data islands in palmprint recognition. There are several clients with different structures deployed in communities, which cannot access the private data of others. The key is to improve the accuracy of each client by generating understandable knowledge and transferring it to each other but without explicitly sharing its private data or model architecture. A public dataset is introduced and several effective communication losses are constructed at both instance level and relation level to help clients to learn from each other. Furthermore, transfer learning is applied to close the gap between private and public data. Extensive experiments are conducted on eighteen constrained and unconstrained palmprint benchmark datasets. The results demonstrate that FedML can outperform other methods by a large margin and obtain promising performance.

Cross-dataset palmprint recognition promotes the convenience and flexibility of palmprint recognition. However, most of the current cross-dataset palmprint recognition methods need to collect the target dataset in advance for model training. They tend to overfit this target dataset and are difficult to generalize to other unknown datasets. In this paper, we propose a novel Palmprint Data and Feature Generation (PDFG) method for a more challenging scenario, Cross-Dataset Palmprint Recognition with Unseen Target dataset (CDPR-UT). Both data-level and feature-level generalization is constructed to improve the adaptability of model to unknown target datasets. A Fourier-based data augmentation method is firstly introduced to generate more training data with new styles. Then several effective losses are constructed at feature level to reduce the shifts between source and augmented datasets and extract adaptive features. Experiments are conducted on multiple palmprint datasets. The results demonstrate that our method is more efficient and robust in dealing with CDPR-UT than other methods. Compared with the baseline, the accuracy is improved by up to 18.20% and the Equal Error Rate (EER) is reduced by up to 12.53%.

As a promising topic in palmprint recognition, cross-dataset palmprint recognition is attracting more and more research interests. In this paper, a more difficult yet realistic scenario is studied, i.e., Single-Source Cross-Dataset Palmprint Recognition with Unseen Target dataset (S2CDPR-UT). It is aimed to generalize a palmprint feature extractor trained only on a single source dataset to multiple unseen target datasets collected by different devices or environments. To combat this challenge, we propose a novel method to improve the generalization of feature extractor for S2CDPR-UT, named Generating stylIzed FeaTures (GIFT). Firstly, the raw features are decoupled into high- and low- frequency components. Then, a feature stylization module is constructed to perturb the mean and variance of low-frequency components to generate more stylized features, which can provided more valuable knowledge. Furthermore, two diversity enhancement and consistency preservation supervisions are introduced at feature level to help to learn the model. The former is aimed to enhance the diversity of stylized features to expand the feature space. Meanwhile, the later is aimed to maintain the semantic consistency to ensure accurate palmprint recognition. Extensive experiments carried out on CASIA Multi-Spectral, XJTU-UP, and MPD palmprint databases show that our GIFT method can achieve significant improvement of performance over other methods. The codes will be released at https://github.com/HuikaiShao/GIFT.

Palmprint recognition has gained increased attention as a novel biometric technology. Nonetheless, it faces a challenge in security as individuals may be able to forge palmprints for malicious purposes. To address this, it is essential to conduct palmprint anti-spoofing detection. Currently, there is a lack of datasets and algorithms in this field. In this paper, we construct a novel, large-scale palmprint attack dataset. Furthermore, we introduce domain generalization into the palmprint anti-spoofing realm. Domain-adversarial training and online triplet mining methods are proposed to enhance generalizability performance for unseen target domains. Experimental results show that compared to baseline, our method achieves superior results on the dataset.

Palmprint recognition is one of the most popular biometric technologies. Recent researches mainly focus on the recognition performance, while pay less attention to the data protection issues. In this paper, we propose an active biometric data protection model for securing palmprint images in transmission or storage scenarios, called Dynamic Random Invisible Watermark Embedding (DRIWE) model. The DRIWE model implicitly embeds a watermark in each original palmprint ROI image, and then separates the embedded watermark from the watermarked image before identification. If the separated watermark is consistent with the original watermark, it indicates that the image is trustworthy and can be used in the subsequent recognition process. Otherwise, it proves that the image has been illegally tampered with. Furthermore, a two-dimensional image information entropy loss is proposed to enhance the generalization of the model to different watermarks. It ensures that the model can always assign enough information to the host image (i.e., original palmprint image) when different watermarks are applied. Thus, it enables the separator to extract the complete watermark from the watermarked image. This greatly enhances the dynamics and randomness of the watermark embedding process and further improves the ability to secure the data. Adequate experiments are conducted on two benchmark palmprint databases. The results show that the proposed DRIWE model has satisfactory attack resistance and strong generalization ability: even if only one watermark is used in training stage, it can be generalized to a dozen of other new watermark images in the testing stage. In addition, the optimal accuracy of the watermarked data is only reduced by 0.07% compared with the original data.

As an effective method of biometrics, palmprint recognition allows the safe identity recognition of humans without spatial and temporal limitations. To build a more robust palmprint recognition system, recent promising Convolutional Neural Networks (CNN) has been incorporated for better palmprint feature extraction and representation. However, the increasing number of palmprint datasets presents us with a cross-domain recognition problem where the upcoming images may come from different imaging conditions compared to the registered palmprints, which will undermine the recognition accuracy significantly. As a supervised approach, the performance of CNN-based model depends on the availability of data and labels from the same domain, which is hard for transferring recognition. To keep the outperforming recognition result of CNN-based models, we propose a novel Regularized Adversarial Domain Adaptative Hashing method (R-ADAH) for cross-domain palmprint recognition based on Deep Hashing Network (DHN). During training, the Maximum Mean Discrepancy (MMD) is incorporated for better adaptive performance. In this scenario, we only train a DHN on the source domain. With the adversarial training, the target network is becoming adaptive to the unlabeled palmprint images with more stable training, unbiased sample gradient and less sensitivity to the hyper-parameter tuning when only domain-specific label is provided. Extensive validation experiments are conducted on benchmark datasets and our self-collected palmprint datasets by mobile phones to test the performance of our model. The results show a promising increase of the recognition performance.

Benefiting from the advantages of safety and reliability, deep learning-based palmprint recognition has attracted widespread attention. However, previous methods are mainly focused on palmprint recognition in a single dataset. In some realistic applications, a certain number of palmprint images collected from multiple devices under different conditions may be available. Due to the existing gaps between different datasets, how to efficiently use them to obtain satisfactory performance is an important and challenging issue. In this paper, we propose a novel Learning with Partners (LWP) framework to improve the multi-source cross-dataset palmprint recognition. Multiple labeled source datasets and an unlabeled dataset are selected as partners to train two feature extractors FS and FT. Firstly, FS is trained as a teacher using labeled source samples to help learn FT. Then, adaptation loss is introduced to constrain the discrepancy between source and target datasets. To alleviate the negative impact of unlabeled target samples on the model, consistency loss including two distance losses are further proposed to correct the misleading in time. Finally, FT can extract adaptive features to match the target with sources. Extensive experiments are conducted on several benchmark palmprint databases and the results demonstrate that our proposed LWP can outperform other comparative baselines by a large margin. The codes are publicly available at http://gr.xjtu.edu.cn/web/bell.

Deep learning-based palmprint recognition algorithms have shown great potential. Most of them are mainly focused on identifying samples from the same dataset. However, they may be not suitable for a more convenient case that the images for training and test are from different datasets, such as collected by embedded terminals and smartphones. Therefore, we propose a novel Joint Pixel and Feature Alignment (JPFA) framework for such cross-dataset palmprint recognition scenarios. Two-stage alignment is applied to obtain adaptive features in source and target datasets. 1) Deep style transfer model is adopted to convert source images into fake images to reduce the dataset gaps and perform data augmentation on pixel level. 2) A new deep domain adaptation model is proposed to extract adaptive features by aligning the dataset-specific distributions of target-source and target-fake pairs on feature level. Adequate experiments are conducted on several benchmarks including constrained and unconstrained palmprint databases. The results demonstrate that our JPFA outperforms other models to achieve the state-of-the-arts. Compared with baseline, the accuracy of cross-dataset identification is improved by up to 28.10% and the Equal Error Rate (EER) of cross-dataset verification is reduced by up to 4.69%. To make our results reproducible, the codes are publicly available at http://gr.xjtu.edu.cn/web/bell/resource.

As a promising topic of biometrics, palmprint recognition helps to effectively verify a person's identity, which is suitable for building a security system. Recent progress has achieved high recognition accuracy in different benchmark datasets due to deep learning. However, these applications are almost implemented in one dataset with iterative training epochs to help neural network generalize. When applied practically where many new users' palmprints registered in sequence, deep learning-based recognition systems cannot avoid the problem of catastrophic forgetting. In this paper, we propose a continual learning framework based on reinforcement learning to dynamically expand the neural network when facing newly registered palmprints without costly retraining or fine-tuning. Experiments on different datasets demonstrate the high adaptability of our model that is promising for solving the forgetting attack of every biometric system.

Palmprint recognition allows accurate identity verification to build a security system. Recently, researchers introduce deep learning to this area that largely improves the recognition accuracy. However, as a supervised approach, its performance relies on availability of data and labels for every registered identity. For large-scale security systems, after image acquisition, we need to check the whole dataset and manually assign labels through comparison, which is a time-consuming task. Besides, labelling some redundant training samples contributes little to the recognition result. In this paper, we introduce an active learning framework to select the best sample set for label assignment. We regard the active learning as a binary classification task and attempt to make the labeled and unlabeled set indistinguishable. Experiments on different datasets demonstrate our model can reduce the annotation cost while achieving comparable recognition performance.

Recently, excellent palmprint recognition algorithms have emerged and achieved satisfactory performance. However, cross-domain palmprint recognition is rarely considered. In this paper, we proposed transfer autoencoder for crossdomain palmprint recognition. Convolutional autoencoders were firstly used to extract low-dimensional features. A discriminator was then introduced to reduce the gap of two domains. The autoencoders and discriminator were alternately trained, and finally the features with the same distribution were extracted. The databases collected from different environments are defined as source and target domains, respectively. Based on the labels in source domain, unsupervised identification of target domain can be achieved. The experiments were performed on 24 cross-domain pairs composed by multispectral database and our self-built uncontrolled databases. The results show that transfer autoencoder can greatly improve cross-domain recognition accuracy, up to 23.26%. At the same time, the accuracy in a single domain can reach over 99% in controlled database.

Nowadays, many efficient palmprint recognition algorithms have emerged. However, previous algorithms can only be used in a single domain. Furthermore, they also require a large amount of labeled data, which is difficult and costly to obtain. In order to solve these problems, we proposed PalmGAN for cross-domain palmprint recognition. Firstly, the labeled fake images were generated to reduce domain gaps, whose styles are similar to the target domain, and at the same time, the identity information remains unchanged. Based on these fake images, supervised Deep Hash Network (DHN) can be trained and directly used for unsupervised identification in the target domain. Moreover, we established semi-uncontrolled and uncontrolled databases, which were collected in uncontrolled environments. Experiments on several popular databases and self-built databases obtained satisfactory performances. PalmGAN can effectively achieve up to 5.08% improvement for cross-domain recognition, and Equal Error Rate (EER) can decrease to 0% for cross-domain recognition between Blue and Green databases.

At present, the fusion of different unimodal biometrics has attracted increasing attention from researchers, who are dedicated to the practical application of biometrics. In this paper, we explored a multi-biometric algorithm that integrates palmprints and dorsal hand veins (DHV). Palmprint recognition has a rather high accuracy and reliability, and the most significant advantage of DHV recognition is the biopsy (Liveness detection). In order to combine the advantages of both and implement the fusion method, deep learning and graph matching were, respectively, introduced to identify palmprint and DHV. Upon using the deep hashing network (DHN), biometric images can be encoded as 128-bit codes. Then, the Hamming distances were used to represent the similarity of two codes. Biometric graph matching (BGM) can obtain three discriminative features for classification. In order to improve the accuracy of open-set recognition, in multi-modal fusion, the score-level fusion of DHN and BGM was performed and authentication was provided by support vector machine (SVM). Furthermore, based on DHN, all four levels of fusion strategies were used for multi-modal recognition of palmprint and DHV. Evaluation experiments and comprehensive comparisons were conducted on various commonly used datasets, and the promising results were obtained in this case where the equal error rates (EERs) of both palmprint recognition and multi-biometrics equal 0, demonstrating the great superiority of DHN in biometric verification.