Query Features Research Articles

A Self-Distillation Embedded Supervised Affinity Attention Model for Few-Shot Segmentation

Few-shot segmentation focuses on the generalization of models to segment unseen object with limited annotated samples. However, existing approaches still face two main challenges. First, huge feature distinction between support and query images causes knowledge transferring barrier, which harms the segmentation performance. Second, limited support prototypes cannot adequately represent features of support objects, hard to guide high-quality query segmentation. To deal with the above two issues, we propose self-distillation embedded supervised affinity attention model to improve the performance of few-shot segmentation task. Specifically, the self-distillation guided prototype module uses self-distillation to align the features of support and query. The supervised affinity attention module generates high-quality query attention map to provide sufficient object information. Extensive experiments prove that our model significantly improves the performance compared to existing methods. Comprehensive ablation experiments and visualization studies also show the significant effect of our method on few-shot segmentation task. On COCO-20i dataset, we achieve new state-of-the-art results. Training code and pretrained models are available at https://github.com/cv516Buaa/SD-AANet.

Gaussian meta-feature balanced aggregation for few-shot synthetic aperture radar target detection

Due to the high mobility and strong concealment characteristics of synthetic aperture radar (SAR) targets, the corresponding SAR datasets exhibit few-shot data properties, and there is a significant lack of research on few-shot target detection methods in the SAR domain. Furthermore, this study is subject to the following limitations: the scarcity of SAR data and significant sample variations make it difficult to control class centers using existing methods, and the learned models tend to be biased towards base classes while easily confusing novel classes with base classes. These limitations hinder the generalization of knowledge from base classes when detecting novel class targets. In this work, we propose a novel few-shot SAR target detection method based on Gaussian meta-feature balanced aggregation (GMFBA), which is based on meta-learning. Specifically, we first propose two novel feature aggregation methods with Gaussian metrics, namely Gaussian projection distribution metric (GPDM) and Gaussian kernel mean embedding metric (GKMEM). By estimating class distribution with variational autoencoders to replace traditional class prototypes, we sample from robust distributions and measure projection Wasserstein distance and Gaussian kernel mean embedding distance with prior distributions, obtaining the best robust support features under the optimal measurement results. Then, based on GPDM and GKMEM, we propose a novel balanced inter-class uncorrelated aggregation (BICUA) method, which extracts support features of each class according to the proportion of samples and aggregates them with query features in a balanced manner, promoting feature representation between different classes and ensuring no interference between features to significantly reduce confusion between base classes and novel classes. Specifically, GMFBA outperforms the state-of-the-art method G-FSOD significantly in all settings, achieving state-of-the-art performance. In contrast, the novel class detection performance of GMFBA has shown an average improvement of 8.56% on split1 and split2 of the SRSDD-v1.0 dataset, and an average improvement of 1.41% on split1 and split2 of the MSAR-1.0 dataset. The code is available at https://github.com/Caltech-Z/GMFBA.

LSQ 2.0: A linked dataset of SPARQL query logs

We present the Linked SPARQL Queries (LSQ) dataset, which currently describes 43.95 million executions of 11.56 million unique SPARQL queries extracted from the logs of 27 different endpoints. The LSQ dataset provides RDF descriptions of each such query, which are indexed in a public LSQ endpoint, allowing interested parties to find queries with the characteristics they require. We begin by describing the use cases envisaged for the LSQ dataset, which include applications for research on common features of queries, for building custom benchmarks, and for designing user interfaces. We then discuss how LSQ has been used in practice since the release of four initial SPARQL logs in 2015. We discuss the model and vocabulary that we use to represent these queries in RDF. We then provide a brief overview of the 27 endpoints from which we extracted queries in terms of the domain to which they pertain and the data they contain. We provide statistics on the queries included from each log, including the number of query executions, unique queries, as well as distributions of queries for a variety of selected characteristics. We finally discuss how the LSQ dataset is hosted and how it can be accessed and leveraged by interested parties for their use cases.

Learning self-target knowledge for few-shot segmentation

Few-shot semantic segmentation uses a few annotated data of a specific class in the support set to segment the target of the same class in the query set. Most existing approaches fail to perform well when there are significant intra-class variances. This paper alleviates the problem by concentrating on mining the query image and using the support set as supplementary information. First, it proposes a Query Prototype Generation Module to generate a query foreground prototype from the query features. Specifically, we use both prototype-level and pixel-level similarity matching to generate two complementary initial prototypes, which we then integrate to create a discriminative query foreground prototype. Second, we propose a Support Auxiliary Refinement Module to further guide the final precise prediction of the query image by leveraging the target category information of the support set through step-by-step mining. Specifically, we generate a query-support mixture prototype based on the support prototype representation obtained using the attention mechanism. Then we generate a support supplement prototype to complement the missing information by encoding over the foreground regions that the query-support mixture prototype fails to segment out. Extensive experiments on PASCAL-5i and COCO-20i demonstrate that our model outperforms the prior works of few-shot segmentation.

Few-shot learning based oral cancer diagnosis using a dual feature extractor prototypical network

A large global health issue is cancer, wherein early diagnosis and treatment have proven to be life-saving. This holds true for oral cancer, thus emphasizing the significance of timely intervention. Deep learning techniques have gained traction in early cancer detection, exhibiting promising outcomes in accurate diagnosis. However, collecting a substantial amount of training data poses a challenge for deep learning models in cancer diagnosis. To address this limitation, this study proposes an oral cancer diagnosis approach based on a few-shot learning framework that circumvents the need for extensive training data. Specifically, a prototypical network is employed to construct a diagnostic model, wherein two feature extractors are utilized to extract prototypical features and query features respectively, departing from the conventional use of a single feature extraction function in prototypical networks. Moreover, a customized loss function is designed for the proposed method. Rigorous experimentation using a histopathological image dataset demonstrates the superior performance of our proposed approach over comparison methods.

Few-shot segmentation using distribution adaption networks

Few-shot segmentation aims to achieve object-specific semantic segmentation in query images by conditioning on specific objects identified within support images. Previous methodologies predominantly relied on evaluating similarities between semantic-level prototypes of support features or comprehensive pixelwise support information and query images to inform predictions for queries. However, a significant domain discrepancy has been identified between the extracted query features and their support counterparts, resulting in a significant misalignment in the distribution of support and query features. As a result, direct analogies between support and query information cannot ensure the precision of query predictions. In this investigation, a distribution adaptation network is introduced, meticulously designed to learn discriminative representations and alleviate the distributional disparities between support features and query features. The adaptation network embeds both support and query features into Reproducing Kernel Hilbert Spaces (RKHS) and optimally aligns their distributions. The distributional gap is effectively bridged through optimal kernel embedding alignment, facilitated by a multiple kernel learning strategy. Extensive experimentation on datasets PASCAL-5i, COCO-20i and FSS-1000 unequivocally demonstrates that the proposed Kernel-Aligned Distribution Network (KADN) achieves state-of-the-art performance.

Directional region-based feature point matching algorithm based on SURF.

Feature point matching is one of the fundamental tasks in binocular vision. It directly affects the accuracy and quality of 3D reconstruction. This study proposes a directional region-based feature point matching algorithm based on the SURF algorithm to improve the accuracy of feature point matching. First, same-name points are selected as the matching reference points in the left and right images. Then, the SURF algorithm is used to extract feature points and construct the SURF feature point descriptors. During the matching process, the location relationship between the query feature point and the reference point in the left image is directed to determine the corresponding matching region in the right image. Then, the matching is completed within this region based on Euclidean distance. Finally, the grid-based motion statistics algorithm is used to eliminate mismatches. Experimental results show that the proposed algorithm can substantially improve the matching accuracy and the number of valid matched points, particularly in the presence of a large amount of noise and interference. It also exhibits good robustness and stability.

Reference Twice: A Simple and Unified Baseline for Few-Shot Instance Segmentation.

Few-Shot Instance Segmentation (FSIS) requires detecting and segmenting novel classes with limited support examples. Existing methods based on Region Proposal Networks (RPNs) face two issues: 1) Overfitting suppresses novel class objects; 2) Dual-branch models require complex spatial correlation strategies to prevent spatial information loss when generating class prototypes. We introduce a unified framework, Reference Twice (RefT), to exploit the relationship between support and query features for FSIS and related tasks. Our three main contributions are: 1) A novel transformer-based baseline that avoids overfitting, offering a new direction for FSIS; 2) Demonstrating that support object queries encode key factors after base training, allowing query features to be enhanced twice at both feature and query levels using simple cross-attention, thus avoiding complex spatial correlation interaction; 3) Introducing a class-enhanced base knowledge distillation loss to address the issue of DETR-like models struggling with incremental settings due to the input projection layer, enabling easy extension to incremental FSIS. Extensive experimental evaluations on the COCO dataset under three FSIS settings demonstrate that our method performs favorably against existing approaches across different shots, e.g., +8.2/ + 9.4 performance gain over state-of-the-art methods with 10/30-shots. Source code and models will be available at this github site.

Prototype Correlation Matching and Class-Relation Reasoning for Few-Shot Medical Image Segmentation.

Few-shot medical image segmentation has achieved great progress in improving accuracy and efficiency of medical analysis in the biomedical imaging field. However, most existing methods cannot explore inter-class relations among base and novel medical classes to reason unseen novel classes. Moreover, the same kind of medical class has large intra-class variations brought by diverse appearances, shapes and scales, thus causing ambiguous visual characterization to degrade generalization performance of these existing methods on unseen novel classes. To address the above challenges, in this paper, we propose a Prototype correlation Matching and Class-relation Reasoning (i.e., PMCR) model. The proposed model can effectively mitigate false pixel correlation matches caused by large intra-class variations while reasoning inter-class relations among different medical classes. Specifically, in order to address false pixel correlation match brought by large intra-class variations, we propose a prototype correlation matching module to mine representative prototypes that can characterize diverse visual information of different appearances well. We aim to explore prototypelevel rather than pixel-level correlation matching between support and query features via optimal transport algorithm to tackle false matches caused by intra-class variations. Meanwhile, in order to explore inter-class relations, we design a class-relation reasoning module to segment unseen novel medical objects via reasoning inter-class relations between base and novel classes. Such inter-class relations can be well propagated to semantic encoding of local query features to improve few-shot segmentation performance. Quantitative comparisons illustrates the large performance improvement of our model over other baseline methods.

Using the power query system for processing and data mining SAP ERP

The article discusses the use of an intelligent Power Query system for processing data from SAP ERP. When used with SAP ERP data, Power Query can greatly simplify the data analysis process, as well as improve the accuracy and reliability of results. The article provides an overview of the main features of Power Query and how they can be used for data processing in conjunction with the SAP ERP system. It highlights the benefits of using Power Query, including speeding up data processing time, improving data accuracy, and the ability to easily exchange and reuse data processing procedures. Using Power Query to process and analyze bank statements via SAP ERP provides many advantages. Power Query's powerful data processing and transformation capabilities enable organizations to analyze their financial data quickly and accurately, making it easier to identify trends, improve decision-making and stimulate business growth. In addition, the integration of Power Query with SAP ERP provides organizations with a centralized representation of financial data in real time, which allows them to make informed decisions and optimize their activities. Using Power Query to analyze bank statements can also help organizations reduce manual data processing and minimize the risk of errors, increasing the accuracy and reliability of financial information. In general, Power Query offers a powerful solution for organizations seeking to optimize financial analysis processes and ensure business success.

Single-Frame Supervision for Spatio-Temporal Video Grounding.

Spatio-Temporal Video Grounding (STVG) aims at localizing the spatio-temporal tube of a specific object in an untrimmed video given a free-form natural language query. As the annotation of tubes is labor intensive, researchers are motivated to explore weakly supervised approaches in recent works, which usually results in significant performance degradation. To achieve a less expensive STVG method with acceptable accuracy, this work investigates the "single-frame supervision" paradigm that requires a single frame labeled with a bounding box within the temporal boundary of the fully supervised counterpart as the supervisory signal. Based on the characteristics of the STVG problem, we propose a Two-Stage Multiple Instance Learning (T-SMILE) method, which creates pseudo labels by expanding the annotated frame to its contextual frames, thereby establishing a fully-supervised problem to facilitate further model training. The innovations of the proposed method are three-folded, including 1) utilizing multiple instance learning to dynamically select instances in positive bags for the recognition of starting and ending timestamps, 2) learning highly discriminative query features by incorporating spatial prior constraints in cross-attention, and 3) designing a curriculum learning-based strategy that iterative assigns dynamic weights to spatial and temporal branches, thereby gradually adapting to the learning branch with larger difficulty. To facilitate future research on this task, we also contribute a large-scale benchmark containing 12,469 videos on complex scenes with single-frame annotation. The extensive experiments on two benchmarks demonstrate that T-SMILE significantly outperforms all weakly-supervised methods. Remarkably, it also performs better than some fully-supervised methods associated with much more annotation labor costs. The dataset and codes are available at https://github.com/qumengxue/T-SMILE.

SCTrans: Self-align and cross-align transformer for few-shot segmentation

Few-shot Semantic Segmentation (FSS) refers to train a segmentation model that can be generalized to novel categories with limited labeled images. One challenge of FSS is spatial inconsistency between support and query images, e.g., appearance and texture. Most existing methods are only committed to utilizing the semantic-level prototypes of support images to guide mask predictions. These methods, nevertheless, only focus on the most discriminate regions of the object rather than holonomic feature representations. Besides, another question exists that the lack of interaction between paired support and query images. In this paper, we propose a self-align and cross-align transformer (SCTrans) to remedy the above limitations. Specifically, we design a feature fusion module (FFM) to incorporate low-level information from the query branch into mid-level semantic features, boosting the semantic representations of query images. In addition, a feature alignment module is designed to bidirectionally propagate semantic information from support to query images conditioned on more representative support and query features, increasing both intra-class similarities and inter-class differences. Extensive experiments on PASCAL-5i and COCO-20i show that our SCTrans significantly advances the state-of-the-art methods.

SQL Injection Detection Using RNN Deep Learning Model

SQL injection attacks are a common type of cyber-attack that exploit vulnerabilities in web applications to access databases through malicious SQL queries. These attacks pose a serious threat to the security and integrity of web applications and their data. The existing methods for detecting SQL injection attacks are based on predefined rules that can be easily circumvented by sophisticated attackers. Therefore, there is a need for a more robust and effective method for detecting SQL injection attacks. In this research, we propose a novel method for detecting SQL injection attacks using recurrent neural networks (RNN), which are a type of deep learning model that can capture the syntax and semantic features of SQL queries. We train an RNN model on a dataset of benign and malicious SQL queries, and use it to classify queries as either benign or malicious. We evaluate our method on a benchmark dataset and compare it with the existing rule-based methods. Our experimental results show that our method achieved high accuracy and outperformed the rule-based methods for detecting SQL injection attacks. Our research contributes to the field of web application security by providing a new and effective solution for protecting web applications from SQL injection attacks using deep learning. Our method has both practical and theoretical implications, as it can be easily integrated into existing web application security frameworks to provide an additional layer of protection against SQL injection attacks, and it can also advance the understanding of how deep learning models can be applied to natural language processing tasks such as SQL query analysis.

Content-based image retrieval based on corel dataset using deep learning

A popular technique for retrieving images from huge and unlabeled image databases are content-based-image-retrieval (CBIR). However, the traditional information retrieval techniques do not satisfy users in terms of time consumption and accuracy. Additionally, the number of images accessible to users are growing due to web development and transmission networks. As the result, huge digital image creation occurs in many places. Therefore, quick access to these huge image databases and retrieving images like a query image from these huge image collections provides significant challenges and the need for an effective technique. Feature extraction and similarity measurement are important for the performance of a CBIR technique. This work proposes a simple but efficient deep-learning framework based on convolutional-neural networks (CNN) for the feature extraction phase in CBIR. The proposed CNN aims to reduce the semantic gap between low-level and high-level features. The similarity measurements are used to compute the distance between the query and database image features. When retrieving the first 10 pictures, an experiment on the Corel-1K dataset showed that the average precision was 0.88 with Euclidean distance, which was a big step up from the state-of-the-art approaches.

Center-Guided Transformer for Panoptic Segmentation

A panoptic segmentation network to predict masks and classes for things and stuff in images is proposed in this work. Recently, panoptic segmentation has been advanced through the combination of the query-based learning and end-to-end learning approaches. Current research focuses on learning queries without distinguishing between thing and stuff classes. We present decoupling query learning to generate effective thing and stuff queries for panoptic segmentation. For this purpose, we adopt different workflows for thing and stuff queries. We design center-guided query selection for thing queries, which focuses on the center regions of individual instances in images, while we set stuff queries as randomly initialized embeddings. Also, we apply a decoupling mask to the self-attention of query features to prevent interactions between things and stuff. In the query selection process, we generate a center heatmap that guides thing query selection. Experimental results demonstrate that the proposed panoptic segmentation network outperforms the state of the art on two panoptic segmentation datasets.

Hierarchical features extraction and data reorganization for code search

According to a natural language query, code search aims to retrieve relevant code snippets from a codebase. Recent works mainly rely on transformer-based pretraining models to measure the matching degree of queries and codes. Compared with works that rely on earlier deep learning methods, such as LSTM and Attention, they can significantly improve the performance of code search tasks. However, the different layers of the transformer-based models have different features that are intuitive and efficient for understanding the semantics of codes and queries but are rarely considered. Moreover, existing methods do not consider further increasing the amount of training data during training to improve the model’s performance.Toward this end, we propose a novel method called HFEDR, which utilizes the hierarchical features of transformer-based models and reorganizes original training data during a training phase. Specifically, we first extract high-level and low-level features of queries and codes from the higher and lower layers of GraphCodeBERT, respectively, achieving multi-view and comprehensive semantic representation. After that, we organize the original training data into hierarchical-uncorrelated feature pairs and then reorganize them into hierarchical-correlated feature pairs, achieving training the model with more data. Finally, we update the model’s parameters using a contrastive training method. We conduct extensive experiments on CodeSearchNet, demonstrating the effectiveness and rationality of our proposed approach.

Contrastive enhancement using latent prototype for few-shot segmentation

Few-shot segmentation is a task that aims to generalize well to segment novel categories in images leveraging only a few annotated samples. Most existing methods adopt the prototype learning architecture, where support prototype vectors are expanded and concatenated with query features to perform conditional segmentation. However, such framework may potentially focus more on query features and neglect the similarity between support and query features. Moreover, it is valuable to mine information and latent categories in the background region to improve method generalization. To address these issues, this paper proposes a contrastive enhancement approach that can mine latent prototypes from background regions and leverage latent classes to raise the utilization of similarity information between prototype and query features. Specifically, a latent prototype sampling module is proposed to generate pseudo-mask and novel prototypes from the region with same category. It is based on feature similarity in high-level features, which is cost-effective for conducting end-to-end learning. This module does not require prior knowledge of object categories and has no strong dependence on clustering numbers like cluster-based method. Besides, a contrastive enhancement module is developed to drive models to provide different predictions with the same query features. It introduces the latent prototypes to regularize the decoder to concentrate more on similarity between support and query features. The proposed modules can be used as auxiliary modules to flexibly integrate into other baselines for better segmentation performance. Extensive experiments show our approach remarkably improves the performance of state-of-the-art methods for 1-shot and 5-shot segmentation, especially outperforming baseline by 5.9% and 7.3% for 5-shot task on Pascal-5i and COCO-20i.

Uncertainty-aware refinement framework for ovarian tumor segmentation in CECT volume.

Ovarian cancer is a highly lethal gynecological disease. Accurate and automated segmentation of ovarian tumors in contrast-enhanced computed tomography (CECT) images is crucial in the radiotherapy treatment of ovarian cancer, enabling radiologists to evaluate cancer progression and develop timely therapeutic plans. However, automatic ovarian tumor segmentation is challenging due to factors such as inhomogeneous background, ambiguous tumor boundaries, and imbalanced foreground-background, all of which contribute to high predictive uncertainty for a segmentationmodel. To tackle these challenges, we propose an uncertainty-aware refinement framework that aims to estimate and refine regions with high predictive uncertainty for accurate ovarian tumor segmentation in CECTimages. To this end, we first employ an approximate Bayesian network to detect coarse regions of interest (ROIs) of both ovarian tumors and uncertain regions. These ROIs allow a subsequent segmentation network to narrow down the search area for tumors and prioritize uncertain regions, resulting in precise segmentation of ovarian tumors. Meanwhile, the framework integrates two guidance modules that learn two implicit functions capable of mapping query features sampled according to their uncertainty to organ or boundary manifolds, guiding the segmentation network to facilitate information encoding of uncertainregions. Firstly, 367 CECT images are collected from the same hospital for experiments. Dice score, Jaccard, Recall, Positive predictive value (PPV), 95% Hausdorff distance (HD95) and Average symmetric surface distance (ASSD) for the testing group of 77 cases are 86.31%, 73.93%, 83.95%, 86.03%, 15.17 mm and 2.57 mm, all of which are significantly better than that of the other state-of-the-art models. And results of visual comparison shows that the compared methods have more mis-segmentation than our method. Furthermore, our method achieves a Dice score that is at least 20% higher than the Dice scores of other compared methods when tumor volumes are less than 20 cm , indicating better recognition ability to small regions by our method. And then, 38 CECT images are collected from another hospital to form an external testing group. Our approach consistently outperform the compared methods significantly, with the external testing group exhibiting substantial improvements across key evaluation metrics: Dice score (83.74%), Jaccard (69.55%), Recall (82.12%), PPV (81.61%), HD95 (12.31mm), and ASSD (2.32mm), robustly establishing its superiorperformance. Experimental results demonstrate that the framework significantly outperforms the compared state-of-the-art methods, with decreased under- or over-segmentation and better small tumor identification. It has the potential for clinicalapplication.

Learn to Few-Shot Segment Remote Sensing Images from Irrelevant Data

Few-shot semantic segmentation (FSS) is committed to segmenting new classes with only a few labels. Generally, FSS assumes that base classes and novel classes belong to the same domain, which limits FSS’s application in a wide range of areas. In particular, since annotation is time-consuming, it is not cost-effective to process remote sensing images using FSS. To address this issue, we designed a feature transformation network (FTNet) for learning to few-shot segment remote sensing images from irrelevant data (FSS-RSI). The main idea is to train networks on irrelevant, already labeled data but inference on remote sensing images. In other words, the training and testing data neither belong to the same domain nor category. The FTNet contains two main modules: a feature transformation module (FTM) and a hierarchical transformer module (HTM). Among them, the FTM transforms features into a domain-agnostic high-level anchor, and the HTM hierarchically enhances matching between support and query features. Moreover, to promote the development of FSS-RSI, we established a new benchmark, which other researchers may use. Our experiments demonstrate that our model outperforms the cutting-edge few-shot semantic segmentation method by 25.39% and 21.31% in the one-shot and five-shot settings, respectively.

Task-Aware Dual-Representation Network for Few-Shot Action Recognition

Few-shot action recognition has attracted increasing attention in recent years, but it remains challenging due to the intrinsic difficulty in learning transferable knowledge to generalize to novel classes by using a few labeled samples. Although some successful progress has been made, most few-shot action recognition methods commonly focus on the global characteristics of samples while ignoring the local characteristics of samples, which results in the weak generalization ability of the model. In this paper, we propose a task-aware dual-representation network (TADRNet) for few-shot action recognition, which learns how to adapt video representations to novel tasks in a meta-learning manner. It mainly includes a global relational graph subnetwork (GRG) and a fine-grained local representation subnetwork (FLR). Our method simultaneously considers both global and local characteristics of samples for few-shot action recognition. From a global perspective, we propose GRG to explore the relations across support-query sample pairs by using the relational graph neural network. To facilitate the few-shot visual learning, we propose a novel hybrid semantic attention module (HSA) for enhancing the discriminability of support and query features. From a local perspective, we utilize FLR to fully exploit the local characteristics of samples, which can improve the classification results obtained by GRG and thus guarantee high classification accuracy. Extensive experiments on four challenging benchmarks show that the proposed TADRNet significantly outperforms a variety of state-of-the-art few-shot action recognition methods.