DTHN: Dual-Transformer Head End-to-End Person Search Network

Person search mainly consists of two submissions,namely Person Detection and Person Re-identification(reID).Existing approaches are primarily based on Faster R-CNN and Convolutional Neural Network(CNN)(e.g.,ResNet).While these structures may detect high-quality bounding boxes,they seem to degrade the performance of re-ID.To address this issue,this paper proposes a Dual-Transformer Head Network(DTHN)for end-to-end person search,which contains two independent Transformer heads,a box head for detecting the bounding box and extracting efficient bounding box feature,and a re-ID head for capturing high-quality re-ID features for the re-ID task.Specifically,after the image goes through the ResNet backbone network to extract features,the Region Proposal Network(RPN)proposes possible bounding boxes.The box head then extracts more efficient features within these bounding boxes for detection.Following this,the re-ID head computes the occluded attention of the features in these bounding boxes and distinguishes them from other persons or backgrounds.Extensive experiments on two widely used benchmark datasets,CUHK-SYSU and PRW,achieve state-of-the-art performance levels,94.9 mAP and 95.3 top-1 scores on the CUHK-SYSU dataset,and 51.6 mAP and 87.6 top-1 scores on the PRW dataset,which demonstrates the advantages of this paper’s approach.The efficiency comparison also shows our method is highly efficient in both time and space.

DAAPS: A Deformable-Attention-Based Anchor-Free Person Search Model

Person Search is a task involving pedestrian detection and person re-identification,aiming to retrieve person images matching a given objective attribute from a large-scale image library.The Person Search models need to understand and capture the detailed features and context information of smaller objects in the image more accurately and comprehensively.The current popular Person Search models,whether end-to-end or two-step,are based on anchor boxes.However,due to the limitations of the anchor itself,the model inevitably has some disadvantages,such as unbalance of positive and negative samples and redundant calculation,which will affect the performance of models.To address the problem of fine-grained understanding of target pedestrians in complex scenes and small sizes,this paper proposes a Deformable-Attention-based Anchor-free Person Search model(DAAPS).Fully Convolutional One-Stage(FCOS),as a classic Anchor-free detector,is chosen as the model’s infrastructure.The DAAPS model is the first to combine the Anchor-free Person Search model with Deformable Attention Mechanism,applied to guide the model adaptively adjust the perceptual.The Deformable Attention Mechanism is used to help the model focus on the critical information and effectively improve the poor accuracy caused by the absence of anchor boxes.The experiment proves the adaptability of the Attention mechanism to the Anchor-free model.Besides,with an improved ResNeXt+network frame,the DAAPS model selects the Triplet-based Online Instance Matching(TOIM)Loss function to achieve a more precise end-to-end Person Search task.Simulation experiments demonstrate that the proposed model has higher accuracy and better robustness than most Person Search models,reaching 95.0%of mean Average Precision(mAP)and 95.6%of Top-1 on the CUHK-SYSU dataset,48.6%of mAP and 84.7%of Top-1 on the Person Re-identification in the Wild(PRW)dataset,respectively.

Ranking of Web Pages in a Personalized Search

The basic idea behind a personalized web search is to deliver search results that are tailored to meet user needs, which is one of the growing concepts in web technologies. The personalized web search presented in this paper is based on exploiting the implicit feedbacks of user satisfaction during her web browsing history to construct a user profile storing the web pages the user is highly interested in. A weight is assigned to each page stored in the user’s profile;this weight reflects the user’s interest in this page. We name this weight the relative rank of the page, since it depends on the user issuing the query. Therefore, the ranking algorithm provided in this paper is based on the principle that;the rank assigned to a page is the addition of two rank values R_rank and A_rank. A_rank is an absolute rank, since it is fixed for all users issuing the same query, it only depends on the link structures of the web and on the keywords of the query. Thus, it could be calculated by the PageRank algorithm suggested by Brin and Page in 1998 and used by the google search engine. While, R_rank is the relative rank, it is calculated by the methods given in this paper which depends mainly on recording implicit measures of user satisfaction during her previous browsing history.

A Personalized Search Model Using Online Social Network Data Based on a Holonic Multiagent System

Personalized search utilizes user preferences to optimize search results,and most existing studies obtain user preferences by analyzing user behaviors in search engines that provide click-through data.However,the behavioral data are noisy because users often clicked some irrelevant documents to find their required information,and the new user cold start issue represents a serious problem,greatly reducing the performance of personalized search.This paper attempts to utilize online social network data to obtain user preferences that can be used to personalize search results,mine the knowledge of user interests,user influence and user relationships from online social networks,and use this knowledge to optimize the results returned by search engines.The proposed model is based on a holonic multiagent system that improves the adaptability and scalability of the model.The experimental results show that utilizing online social network data to implement personalized search is feasible and that online social network data are significant for personalized search.

Community-Aware Resource Profiling for Personalized Search in Folksonomy

In recent years, there is a fast proliferation of collaborative tagging （a.k.a. folksonomy） systems in Web 2.0 communities. With the increasingly large amount of data, how to assist users in searching their interested resources by utilizing these semantic tags becomes a crucial problem. Collaborative tagging systems provide an environment for users to annotate resources, and most users give annotations according to their perspectives or feelings. However, users may have different perspectives or feelings on resources, e.g., some of them may share similar perspectives yet have a conflict with others. Thus, modeling the profile of a resource based on tags given by all users who have annotated the resource is neither suitable nor reasonable. We propose, to tackle this problem in this paper, a community-aware approach to constructing resource profiles via social filtering. In order to discover user communities, three different strategies are devised and discussed. Moreover, we present a personalized search approach by combining a switching fusion method and a revised needs-relevance function, to optimize personalized resources ranking based on user preferences and user issued query. We conduct experiments on a collected real life dataset by comparing the performance of our proposed approach and baseline methods. The experimental results verify our observations and effectiveness of proposed method.

A comprehensive review from hyperlink to intelligent technologies based personalized search systems

In the present era of big data,web page searching and ranking in an efficient manner on the World Wide Web to satisfy the specific search needs of the modern user is undoubtedly a major challenge for search engines.Even though a large number of web search techniques have been developed,some problems still exist while searching with generic search engines as none of the search engines can index the entire web.The issue is not just the volume but also the relevance concerning the user’s requirements.Moreover,if the search query is partially incomplete or is ambiguous,then most of the modern search engines tend to return the result by interpreting all possible meanings of the query.Concerning search quality,more than half of the retrieved web pages have been reported to be irrelevant.Hence web search personalization is required to retrieve search results while incorporating the user’s interests.In the proposed research work we have highlighted the strengths and weakness of various studies as proposed in the literature for web search personalization by carrying out a detailed comparison among them.The in-depth comparative study with baselines leads to the recommendation of Intelligent Meta Search System(IMSS)and Advanced Cluster Vector Page Ranking(ACVPR)algorithm as one of the best approaches as proposed in the literature for web search personalization.Furthermore,the detailed discussion about the comparative analysis of all categories gives new opportunities to think in different research areas.

AOL4PS:A Large-scale Data Set for Personalized Search

Personalized search is a promising way to improve the quality of Websearch,and it has attracted much attention from both academic and industrial communities.Much of the current related research is based on commercial search engine data,which can not be released publicly for such reasons as privacy protection and information security.This leads to a serious lack of accessible public data sets in this field.The few publicly available data sets have not become widely used in academia because of the complexity of the processing process required to study personalized search methods.The lack of data sets together with the difficulties of data processing has brought obstacles to fair comparison and evaluation of personalized search models.In this paper,we constructed a large-scale data set AOL4 PS to evaluate personalized search methods,collected and processed from AOL query logs.We present the complete and detailed data processing and construction process.Specifically,to address the challenges of processing time and storage space demands brought by massive data volumes,we optimized the process of data set construction and proposed an improved BM25 algorithm.Experiments are performed on AOL4 PS with some classic and state-of-the-art personalized search methods,and the experiment results demonstrate that AOL4 PS can measure the effect of personalized search models.