Target Person Research Articles

Text-and-Image Learning Transformer for Cross-modal Person Re-identification

Text-based person re-identification aims to find the target person from a large pedestrian gallery with the given natural language description. Previous works mainly focus on embedding salient textual and visual representations in a common latent space by utilizing the dual-path structure or parameter-shared network. However, they still lack the ability to effectively extract fine-grained unimodal features as well as fuse the cross-modal data, leading to the increase of misaligned cases. To settle these issues, we propose a text-and-image implicit learning Transformer (TILT) to eliminate textual anisotropy and enhance the cross-modal alignment from both domains based on the bi-direction multimodal encoders. Specifically, we apply the pre-trained multimodal embedding module to overcome the unimodal anisotropy problem with contrastive learning, and map fine-grained features with dual encoder in bidirectional masking. Then, we design the cross-modal interaction encoder to comprehensively mine implicit cross-modal relations by reconstructing masked tokens, and fuse rich multi-modal knowledge in a common space. In addition, the cross-modal similarity matching module is proposed to optimize the intra-domain classification and decrease the inter-domain divergence. Extensive experiments are conducted on three public benchmarks CUHK-PEDES, ICFG-PEDES and RSTPReid to verify the effectiveness of our proposed framework. And results prove that our model outperforms state-of-the-art methods on all metrics.

A Person Re-Identification Method for Fusing Convolutional Attention and Transformer Architecture

Rapid and accurate localization of target person via person re-identification is one of the important methods in intelligent security systems,but faces great challenges due to complex environmental changes. In order to build a model applicable to various complex scenes, deep learning-based person re-identification models have become a key research direction for researchers. The existing convolutional neural network-based or transformer-based deep learning models are still have certain shortcomings in the extraction of local details and are not able to cope well with realistic complex scenes. To address this problem, a method fusing convolutional neural network and transformer (FCAT) is proposed to enhance the transformer's attention to local detail information. This method mainly improves the transformer's ability to extract local detail features indirectly by embedding convolutional space attention and channel attention respectively to enhance the attention to important regions and important channel features in the image. Comparative ablation experiments on three publicly available person re-identification datasets demonstrate that FCAT achieves comparable results to existing methods on non-occluded datasets, and significantly improves performance on occluded datasets. The proposed model is also has a lightweight effect, and the inference speed is improved without increasing the amount of outer computation and model parameters.

Spotting missing or wanted people: racial biases in prospective person memory

Searching for missing or wanted people is a crucial task in our society. Previous work on prospective person memory (PPM) has demonstrated that performance on this type of search task is worse relative to standard prospective memory tasks. Importantly, this process may be further affected by the race of the missing person, yet this has never been tested in laboratory settings. To test the effects of race on PPM, a convenience sample consisting primarily of self-identified Caucasian participants was asked to search for either a Caucasian or an Indian target person while judging the orientation of different Caucasian and Indian faces. Although the tasks were otherwise identical, 89% of Caucasian PPM targets were found while only 53% of Indian targets were found. Furthermore, relative to a control group with no PPM requirements, participants were slower and more error-prone when judging Indian faces relative to White faces, particularly if they were searching for an Indian face. We interpret these results as revealing other-race effects in prospective person memory, highlighting race as a critical factor for finding missing people. Importantly, this also emphasizes the need for real-world search efforts to factor in difficulty differences when people monitor for missing/wanted people from their own or different racial backgrounds. For example, media coverage of missing persons cases could perhaps be distributed more equitably by considering whether the missing person is from a racial minority in that region.

LQR-based control strategy for improving human-robot companionship and natural obstacle avoidance

In the dynamic and unstructured environment of human–robot symbiosis, companion robots require natural human–robot interaction and autonomous intelligence through multimodal information fusion to achieve effective collaboration. Nevertheless, the control precision and coordination of the accompanying actions are not satisfactory in practical applications. This is primarily attributed to the difficulties in the motion coordination between the accompanying target and the mobile robot. This paper proposes a companion control strategy based on the Linear Quadratic Regulator (LQR) to enhance the coordination and precision of robot companion tasks. This method enables the robot to adapt to sudden changes in the companion target’s motion. Besides, the robot could smoothly avoid obstacles during the companion process. Firstly, a human–robot companion interaction model based on nonholonomic constraints is developed to determine the relative position and orientation between the robot and the companion target. Then, an LQR-based companion controller incorporating behavioral dynamics is introduced to simultaneously avoid obstacles and track the companion target’s direction and velocity. Finally, various simulations and real-world human–robot companion experiments are conducted to regulate the relative position, orientation, and velocity between the target object and the robot platform. Experimental results demonstrate the superiority of this approach over conventional control algorithms in terms of control distance and directional errors throughout system operation. The proposed LQR-based control strategy ensures coordinated and consistent motion with target persons in social companion scenarios.

Leveraging Social Engineering Techniques for Ethical Purposes: An Approach to Develop Fake Android App for Collecting Valuable Data Discreetly

Social engineering techniques are often viewed negatively due to their association with deceptive practices. However, these techniques can also be utilized ethically, as many cybersecurity professionals do, particularly when evaluating vulnerabilities and testing security defenses. This paper presents EDC (Ethical Data Collector), an Android application that utilizes social engineering techniques to discreetly collect valuable data from an Android device for ethical purposes. EDC employs deception through a simulated UI (fake) to engage the target for a period, while secretly collecting data such as device information, active phone number, and images in the background, then sending them to a designated server via the internet. The researcher argues that EDC could help identify inexperienced cybercriminals or extortionists without complex efforts or significant cost, provided that its capabilities are judiciously utilized and subject to proper controls and oversight. EDC's development methodology emphasizes understanding the target's personality, predilections, and preferences to tailor the app experience as required for attracting the target to install and run the application. The paper describes the core functions and workflows for collecting and sending data. Additionally, permissions handling has been addressed as being critical for enabling EDC to collect the required information. Testing on Android emulators demonstrated that the EDC's APK file size is 4 MB, and data collection and transmission processes functioned as intended across various Android versions. The minimum SDK version required to run EDC is level 16. The total estimated time to complete the fake UI process (cumulative user engagement time) is 57-60 seconds, where each activity takes 8 seconds based on the adopted development approach.

KF-VTON: Keypoints-Driven Flow Based Virtual Try-On Network

Image-based virtual try-on aims to fit a target garment to a reference person. Most existing methods are limited to solving the Garment-To-Person (G2P) try-on task that transfers a garment from a clean product image to the reference person and do not consider the Person-To-Person (P2P) try-on task that transfers a garment from a clothed person image to the reference person, which limits the practical applicability. The P2P try-on task is more challenging due to spatial discrepancies caused by different poses, body shapes, and views between the reference person and the target person. To address this issue, we propose a novel Keypoints-Driven Flow Based Virtual Try-On Network (KF-VTON) for handling both the G2P and P2P try-on tasks. Our KF-VTON has two key innovations: (1) We propose a new keypoints-driven flow based deformation model to warp the garment. This model establishes spatial correspondences between the target garment and reference person by combining the robustness of Thin-plate Spline (TPS) based deformation and the flexibility of appearance flow based deformation. (2) We investigate a powerful Context-aware Spatially Adaptive Normalization (CSAN) generative module to synthesize the final try-on image. Particularly, CSAN integrates rich contextual information with semantic parsing guidance to properly infer unobserved garment appearances. Extensive experiments demonstrate that our KF-VTON is capable of producing photo-realistic and high-fidelity try-on results for the G2P as well as P2P try-on tasks and surpasses previous state-of-the-art methods both quantitatively and qualitatively. Our code is available at https://github.com/OIUIU/KF-VTON .

Prosocial behaviour enhances evaluation of physical beauty.

Ten studies (N = 4192) demonstrated that individuals depicted as prosocial were judged to be more physically beautiful. This evaluation of prosocial individuals as more beautiful is influenced by a motivation to be associated with prosocial others. This phenomenon was observed in real-world settings (Study 1) and applied to both men and women, both as targets and observers (Studies 2a-2b). The effect persisted in scenarios where participants imagined the target without any visual aid (Study 2c) and extended beyond metaphorical interpretations of beauty (Study 3). The effect weakened when prosocial behaviour was an isolated incident, not indicative of the target's prosocial personality (Study 4). The influence of prosociality on beauty evaluations surpassed that of other positive traits such as intelligence or humour (Study 5) and remained significant despite physical imperfections in the target's appearance (Study 6). The effect diminished insituations where forming a relationship was not feasible, thus supporting the motivated cognition rationale (Studies 7-8). These findings highlight the substantial role of prosocial behaviour in influencing evaluations of physical beauty, a crucial element in social interactions and relationship formation, often outweighing other attributes typically linked to physical appearance evaluation.

FICE: Text-conditioned fashion-image editing with guided GAN inversion

Fashion-image editing is a challenging computer-vision task where the goal is to incorporate selected apparel into a given input image. Most existing techniques, known as Virtual Try-On methods, deal with this task by first selecting an example image of the desired apparel and then transferring the clothing onto the target person. Conversely, in this paper, we consider editing fashion images with text descriptions. Such an approach has several advantages over example-based virtual try-on techniques: (i) it does not require an image of the target fashion item, and (ii) it allows the expression of a wide variety of visual concepts through the use of natural language. Existing image-editing methods that work with language inputs are heavily constrained by their requirement for training sets with rich attribute annotations or they are only able to handle simple text descriptions. We address these constraints by proposing a novel text-conditioned editing model called FICE (Fashion Image CLIP Editing) that is capable of handling a wide variety of diverse text descriptions to guide the editing procedure. Specifically, with FICE, we extend the common GAN-inversion process by including semantic, pose-related, and image-level constraints when generating images. We leverage the capabilities of the CLIP model to enforce the text-provided semantics, due to its impressive image–text association capabilities. We furthermore propose a latent-code regularization technique that provides the means to better control the fidelity of the synthesized images. We validate the FICE through rigorous experiments on a combination of VITON images and Fashion-Gen text descriptions and in comparison with several state-of-the-art, text-conditioned, image-editing approaches. Experimental results demonstrate that the FICE generates very realistic fashion images and leads to better editing than existing, competing approaches. The source code is publicly available from: https://github.com/MartinPernus/FICE.

Cybercrime target exposure, suitability, personality, and victimization: A longitudinal approach

This study utilizes the routine activity theory as a framework to examine if workplace social media usage, compulsive internet use, online aggression and personality traits are related to an elevated risk of cybervictimization. We leveraged 7-wave longitudinal survey data from the Finnish working population (N = 650) and employed multilevel, mixed-effects logistic regression to analyze fixed effects of in-between variables. Our findings reveal that target exposure, suitability and the personality trait of openness had a positive relationship with cybervictimization. We observed fluctuations in cybervictimization across the observation points, although the temporal pattern did not follow a linear trajectory. The results underscore the importance of longitudinal studies on victimization and advocate additional research on cybervictimization within the working population. This study also emphasizes the need for the integration of established theories to augment our comprehension of the fundamental factors influencing cybervictimization.

Depth Matters: Spatial Proximity-based Gaze Cone Generation for Gaze Following in Wild

Gaze following aims to predict where a person is looking in a scene. Existing methods tend to prioritize traditional 2D RGB visual cues or require burdensome prior knowledge and extra expensive datasets annotated in 3D coordinate systems to train specialized modules to enhance scene modeling. In this work, we introduce a novel framework deployed on a simple ResNet backbone, which exclusively uses image and depth maps to mimic human visual preferences and realize 3D-like depth perception. We first leverage depth maps to formulate spatial-based proximity information regarding the objects with the target person. This process sharpens the focus of the gaze cone on the specific region of interest pertaining to the target while diminishing the impact of surrounding distractions. To capture the diverse dependence of scene context on the saliency gaze cone, we then introduce a learnable grid-level regularized attention that anticipates coarse-grained regions of interest, thereby refining the mapping of the saliency feature to pixel-level heatmaps. This allows our model to better account for individual differences when predicting others’ gaze locations. Finally, we employ the KL-divergence loss to super the grid-level regularized attention, which combines the gaze direction, heatmap regression, and in/out classification losses, providing comprehensive supervision for model optimization. Experimental results on two publicly available datasets demonstrate the comparable performance of our model with less help of modal information. Quantitative visualization results further validate the interpretability of our method. The source code will be available at https://github.com/VUT-HFUT/DepthMatters .

Journey to the Grave

Signage is an aid to wayfinding and individual orientation in both organisations and everyday life. It aims to respond to users’ needs. Whether or not it actually succeeds in this can only be empirically verified by an evaluation that is tailored to the context in question. Spacious cemeteries are a particularly interesting case, as they are a place both of mourning and of relaxation. When visitors in a fragile emotional state want to find their way to a grave, they have to be able to depend on particularly effective signage. The present study of a pilot project at the Zurich Sihlfeld Cemetery in 2022/2023 uses a “shadowing” methodology in such a context for the first-ever time. This is because other common approaches to evaluating signage are inappropriate here for ethical reasons – whether these be surveys, giving test subjects specific search tasks, or using eye-tracking. We observed and assessed 49 target persons across all the segments of the cemetery in their general orientation behaviour, the degree to which they consulted the signage offered, and their use of other aids on their way to the burial in question. We used our observations to analyse deviations from the ideal access routes; our photographic records provide us with a basis for further optimisation measures.

Predicting Workers' Stress: Application of a High-Performance Algorithm Using Working-Style Characteristics.

Work characteristics, such as teleworking rate, have been studied in relation to stress. However, the use of work-related data to improve a high-performance stress prediction model that suits an individual's lifestyle has not been evaluated. This study aims to develop a novel, high-performance algorithm to predict an employee's stress among a group of employees with similar working characteristics. This prospective observational study evaluated participants' responses to web‑based questionnaires, including attendance records and data collected using a wearable device. Data spanning 12 weeks (between January 17, 2022, and April 10, 2022) were collected from 194 Shionogi Group employees. Participants wore the Fitbit Charge 4 wearable device, which collected data on daily sleep, activity, and heart rate. Daily work shift data included details of working hours. Weekly questionnaire responses included the K6 questionnaire for depression/anxiety, a behavioral questionnaire, and the number of days lunch was missed. The proposed prediction model used a neighborhood cluster (N=20) with working-style characteristics similar to those of the prediction target person. Data from the previous week predicted stress levels the following week. Three models were compared by selecting appropriate training data: (1) single model, (2) proposed method 1, and (3) proposed method 2. Shapley Additive Explanations (SHAP) were calculated for the top 10 extracted features from the Extreme Gradient Boosting (XGBoost) model to evaluate the amount and contribution direction categorized by teleworking rates (mean): low: <0.2 (more than 4 days/week in office), middle: 0.2 to <0.6 (2 to 4 days/week in office), and high: ≥0.6 (less than 2 days/week in office). Data from 190 participants were used, with a teleworking rate ranging from 0% to 79%. The area under the curve (AUC) of the proposed method 2 was 0.84 (true positive vs false positive: 0.77 vs 0.26). Among participants with low teleworking rates, most features extracted were related to sleep, followed by activity and work. Among participants with high teleworking rates, most features were related to activity, followed by sleep and work. SHAP analysis showed that for participants with high teleworking rates, skipping lunch, working more/less than scheduled, higher fluctuations in heart rate, and lower mean sleep duration contributed to stress. In participants with low teleworking rates, coming too early or late to work (before/after 9 AM), a higher/lower than mean heart rate, lower fluctuations in heart rate, and burning more/fewer calories than normal contributed to stress. Forming a neighborhood cluster with similar working styles based on teleworking rates and using it as training data improved the prediction performance. The validity of the neighborhood cluster approach is indicated by differences in the contributing features and their contribution directions among teleworking levels. UMIN UMIN000046394; https://www.umin.ac.jp/ctr/index.htm.

Human-following task without a prior map

Purpose The human-following task is a fundamental function in human–robot collaboration. It requires a robot to recognize and locate a target person, plan a path and avoid obstacles. To enhance the applicability of the human-following task in various scenarios, it should not rely on a prior map. This paper aims to introduce a human-following method that meets these requirements. Design/methodology/approach For the identification and localization of the target person (ILTP), this paper proposes an approach that integrates data from a camera, a light detection and ranging (LiDAR) and a ultra-wideband (UWB) anchor. For path planning and obstacle avoidance, a modified timed-elastic-bands (TEB) algorithm is introduced. Findings Compared to the UWB-only method, where only UWB is used to locate the target person, the proposed ILTP method in this paper reduces the localization error by 41.82%. Experimental results demonstrate the effectiveness of the ILTP and the modified TEB method under various challenging conditions. Such as crowded environments, multiple obstacles, the target person being occluded and the target person moving out of the robot’s field of view. The complete experimental videos are available for viewing on https://youtu.be/ZKbrNE1sePM. Originality/value This paper offers a novel solution for human-following tasks. The proposed ILTP method can recognize the target person among multiple individuals, determine whether the target person is lost and publish the target person’s position at a frequency of 20 Hz. The modified TEB algorithm does not rely on a prior map. It can plan paths and avoid obstacles effectively.

Verbal cues in omission lies: The effect of informing sources about the essential part of the event

AbstractPeople sometimes lie by omitting information. The information lie tellers then report could be entirely truthful. We examined whether the truthful information that lie tellers report in omission lies contains verbal cues indicating that the person is lying. We made a distinction between (i) essential information (events surrounding the omission) and non‐essential information (the rest); and (ii) made a distinction between informing or not informing participants about the key event they witnessed. Participants followed a target person. Truth tellers reported all activities truthfully; lie tellers omitted the key event. Participants were or were not informed what this key event was. In the analyses we discarded the information truth tellers reported about the key event lie tellers omitted. Truth tellers reported more external and contextual details, more complications and fewer common knowledge details and self‐handicapping strategies than lie tellers, but only when discussing essential information. Being informed had no effect.

The effect of prepaid incentives on panelists’ response across survey modes in a sequential mixed-mode design

AbstractWhile the use of prepaid incentives and data collection in a sequential mixed-mode survey design is standard in cross-sectional and longitudinal studies, little is known about the interaction of these strategies and how this affects target persons’ survey participation. Therefore, based on a panel study, this study investigates whether such unconditional prepayments—monetary or non-monetary incentives—increase the efficiency of a sequential mixed-mode survey design in regard to boosted response rates and speed of return after receipt of an invitation to participate in a survey. While the survey design is always identical, the different incentives vary across the waves considered. Different prepayments, such as cash or in-kind incentives, have different effects on invitees’ participation in different survey modes. In particular, non-monetary incentives seem to work only in a sequential mixed-mode design, while cash, as a universal medium, always works in the expected way, across different survey modes. In sum, the study finds that when a single sequential mixed-mode design is used across panel waves the overall response rates are rather similar, even when different incentives are provided.

Concurring Emotions, Affective Empathy, and Phenomenal Understanding

According to an optimistic view, affective empathy is a route to knowledge of what it is like to be in the target person’s state (“phenomenal knowledge”). Roughly, the idea is that the empathizer gains this knowledge by means of empathically experiencing the target’s emotional state. The literature on affective empathy, however, often draws a simplified picture according to which the target feels only a single emotion at a time. Co-occurring emotions (“concurrent emotions”) are rarely considered. This is problematic, because concurring emotions seem to support a sceptical view according to which we cannot gain phenomenal knowledge of the target person’s state by means of affective empathy. The sceptic concludes that attaining the epistemic goal of affective empathy is difficult, in practice often impossible. I accept the sceptic’s premises, but reject the conclusion, because of the argument’s unjustified, hidden premise: that the epistemic goal of affective empathy is phenomenal knowledge. I argue that the epistemic goal of affective empathy is phenomenal understanding, not knowledge. Attention to the under-explored phenomenon of concurring emotions clarifies why this is important. I argue that this is the decisive epistemic progress in everyday cases of phenomenal understanding of another person.

The Significance of Meta-Emotions for Reflecting on Ourselves and Others

Meta-emotions are emotions about emotions, such as, for example, shame about anger. An important subset of meta-emotions constitutes a special case of co-experienced emotions, in which one emotion is directed at another emotion experienced by a subject at the same time. By making us reflectively aware of our own first-order emotions and suggesting an evaluation of them, meta-emotions enable reflection on our own emotional sensibility. In this paper, I explore the roles of meta-emotions in the context of affective empathy. I show that, whereas meta-emotions on the side of the target person increase the complexity of empathy’s success conditions, meta-emotions on the side of the empathiser expand the role of meta-emotions for self-reflection and give rise to parallel phenomena of other-directed reflection. Exploring the roles of meta-emotions in the context of affective empathy allows us to better understand the significance of meta-emotions, as well as the functions of affective empathy.

FA-VTON: A Feature Alignment-Based Model for Virtual Try-On

The virtual try-on technology based on 2D images aims to seamlessly transfer provided garments onto target person images. Prior methods mainly concentrated on warping garments and generating images, overlooking the influence of feature alignment on the try-on results. In this study, we initially analyze the distortions present by existing methods and elucidate the critical role of feature alignment in the extraction stage. Building on this, we propose a novel feature alignment-based model (FA-VTON). Specifically, FA-VTON aligns the upsampled higher-level features from both person and garment images to acquire precise boundary information, which serves as guidance for subsequent garment warping. Concurrently, the Efficient Channel Attention mechanism (ECA) is introduced to generate the final result in the try-on generation module. This mechanism enables adaptive adjustment of channel feature weights to extract important features and reduce artifact generation. Furthermore, to make the student network focus on salient regions of each channel, we utilize channel-wise distillation (CWD) to minimize the Kullback–Leibler (KL) divergence between the channel probability maps of the two networks. The experiments show that our model achieves better results in both qualitative and quantitative analyses compared to current methods on the popular virtual try-on datasets.

The Associations Among Observers' Openness to Experience and Agreeableness With Social Distance: The Moderating Role of Disability Type.

Despite abundant literature on personality and stigma, the role of disability type in this relationship has remained unaddressed. In the current study, we examined whether the relationship between observers' openness to experience and agreeableness on the one hand, and social distance on the other, was moderated by the target person's type of disability (psychiatric vs. physical). One hundred thirty-nine participants were randomly assigned to complete a social distance questionnaire referring to a vignette of a person in three conditions (physical disability, psychiatric disability, and control). A main effect of openness on social distance was found. Additionally, we found an interaction effect of agreeableness and the type of disability. Namely, the relationship between agreeableness and social distance was significant only in the physical disability condition but not in the other two conditions. To conclude, the current study emphasizes the role of personality traits in social distance toward individuals with disabilities.

Differentiation in social perception: Why later-encountered individuals are described more negatively.

According to the cognitive-ecological model of social perception, biases toward individuals can arise as by-products of cognitive principles that interact with the information ecology. The present work tested whether negatively biased person descriptions occur as by-products of cognitive differentiation. Later-encountered persons are described by their distinct attributes that differentiate them from earlier-encountered persons. Because distinct attributes tend to be negative, serial person descriptions should become increasingly negative. We found our predictions confirmed in six studies. In Study 1, descriptions of representatively sampled persons became increasingly distinct and negative with increasing serial positions of the target person. Study 2 eliminated this pattern of results by instructing perceivers to assimilate rather than differentiate a series of targets. Study 3 generalized the pattern from one-word descriptions of still photos of targets to multisentence descriptions of videos of targets. In line with the cognitive-ecological model, Studies 4-5b found that the relation between serial position and negativity was amplified among targets with similar positive attributes, zero among targets with distinct positive or negative attributes, and reversed among similar negative targets. Study 6 returned to representatively sampled targets and generalized the serial position-negativity effect from descriptions of the targets to overall evaluations of them. In sum, the present research provides strong evidence for the explanatory power of the cognitive-ecological model of social perception. We discuss theoretical and practical implications. It may pay off to appear early in an evaluation sequence. (PsycInfo Database Record (c) 2024 APA, all rights reserved).