Search results

Single-shot multi-category clothing recognition and retrieval play a crucial role in online searching and offline settlement scenarios. Existing clothing recognition methods based on RGBD clothing images often suffer from high-dimensional feature representations, leading to compromised performance and efficiency.

To address this issue, this paper proposes a novel method called Manifold Embedded Discriminative Feature Selection (MEDFS) to select global and local features, thereby reducing the dimensionality of the feature representation and improving performance. Specifically, by combining three global features and three local features, a low-dimensional embedding is constructed to capture the correlations between features and categories. The MEDFS method designs an optimization framework utilizing manifold mapping and sparse regularization to achieve feature selection. The optimization objective is solved using an alternating iterative strategy, ensuring convergence.

Empirical studies conducted on a publicly available RGBD clothing image dataset demonstrate that the proposed MEDFS method achieves highly competitive clothing classification performance while maintaining efficiency in clothing recognition and retrieval.

This paper introduces a novel approach for multi-category clothing recognition and retrieval, incorporating the selection of global and local features. The proposed method holds potential for practical applications in real-world clothing scenarios.

Automatic DNA-binding protein (DNA-BP) classification is now an essential proteomic technology. Unfortunately, many systems reported in the literature are tested on only one or two datasets/tasks. The purpose of this study is to create the most optimal and universal system for DNA-BP classification, one that performs competitively across several DNA-BP classification tasks.

Efficient DNA-BP classifier systems require the discovery of powerful protein representations and feature extraction methods. Experiments were performed that combined and compared descriptors extracted from state-of-the-art matrix/image protein representations. These descriptors were trained on separate support vector machines (SVMs) and evaluated. Convolutional neural networks with different parameter settings were fine-tuned on two matrix representations of proteins. Decisions were fused with the SVMs using the weighted sum rule and evaluated to experimentally derive the most powerful general-purpose DNA-BP classifier system.

The best ensemble proposed here produced comparable, if not superior, classification results on a broad and fair comparison with the literature across four different datasets representing a variety of DNA-BP classification tasks, thereby demonstrating both the power and generalizability of the proposed system.

Most DNA-BP methods proposed in the literature are only validated on one (rarely two) datasets/tasks. In this work, the authors report the performance of our general-purpose DNA-BP system on four datasets representing different DNA-BP classification tasks. The excellent results of the proposed best classifier system demonstrate the power of the proposed approach. These results can now be used for baseline comparisons by other researchers in the field.

Recent archiving and curatorial practices took advantage of the advancement in digital technologies, creating immersive and interactive experiences to emphasize the plurality of memory materials, encourage personalized sense-making and extract, manage and share the ever-growing surrounding knowledge. Audiovisual (AV) content, with its growing importance and popularity, is less explored on that end than texts and images. This paper examines the trend of datafication in AV archives and answers the critical question, “What to extract from AV materials and why?”.

This study roots in a comprehensive state-of-the-art review of digital methods and curatorial practices in AV archives. The thinking model for mapping AV archive data to purposes is based on pre-existing models for understanding multimedia content and metadata standards.

The thinking model connects AV content descriptors (data perspective) and purposes (curatorial perspective) and provides a theoretical map of how information extracted from AV archives should be fused and embedded for memory institutions. The model is constructed by looking into the three broad dimensions of audiovisual content – archival, affective and aesthetic, social and historical.

This paper contributes uniquely to the intersection of computational archives, audiovisual content and public sense-making experiences. It provides updates and insights to work towards datafied AV archives and cope with the increasing needs in the sense-making end using AV archives.

The purpose of this study is examined whether coproduction appears spontaneously in the literature on public sector innovation and governance, the citizens’ role in coproduction and the implication of citizens’ participation in the governance of innovation networks.

The review complied with preferred reporting items for systematic reviews and meta-analyses (PRISMA) protocol. The search was performed in the Ebsco, Scopus and WOS databases. The authors analyzed 47 papers published from 2017 to 2022. Thematic and content analysis were adopted, supported by MAXQDA.

The papers recognize the importance of the citizens in public innovation. However, only 20% discuss coproduction, evidencing the predominance of governance concepts related to interorganizational collaborations – but not necessarily to citizen engagement. The authors also verified the existence of polysemy regarding the concept of governance associated with public innovation, predominating the term “collaborative governance.”

The small emphasis on “co-production” may result from the search strategy, which deliberately did not include it as a descriptor, considering the research purpose. One can consider this choice a limitation.

Considering collaborative governance as a governing arrangement where public agencies directly engage nonstate stakeholders in a collective decision-making process that is formal, consensus-oriented and deliberative (Ansell and Gash, 2007), the forum where the citizen is supposed to be engaged should be initiated by public agencies or institutions and formally organized, as suggested by Österberg and Qvist (2020) and Campomori and Casula (2022). These notions can be useful for public managers concerning their role and how the forums structure should be to promote collaboration and the presence of innovation assets needed to make the process fruitful (Crosby et al., 2017).

Despite the collaborative nature of public innovation, the need for adequate governance characteristics, and the importance of citizens in the innovative process, most studies generically address collaborative relationships, focusing on interorganizational collaboration, with little focus on specific actors such as citizens in the governance of public innovation. Thus, it is assumed that the literature that discusses public innovation and governance includes the discussion of coproduction. The originality and contribution of this study is to verify this assumption.

Regarding industrial sports products, there is sometimes a dual sport and health meaning intended by designers. Appearances of sport products are often quite opposite to health products. Design choices made by designers can thus be misunderstood by users. This paper aims to deeper understand the perception gap between designers and users within earlier stages of the design process to limit this confusion and help designers.

The authors propose an approach to help designers defining the perception of a new dual and hybrid product field. The first step is to collect designers’ perception through interviews combined with card sorting. The second step is to compare the perception of designers with that of users. Comparisons are based on an agreement measure.

The approach provides a first step to evaluate the perception of a dual hybrid product field. It allows designers to extract trends and perceptions to be considered for the design of products, to consolidate and confirm their intuitions regarding the intended dual meaning.

The main contribution of this paper is to evaluate the perception of a new and non-defined hybrid product field presenting a duality in appearance. This approach can be used by designers either to identify trends to be considered, reinforce the intended meaning, or validate their intuitions while designing products with dual meanings before.

This paper aims to propose a robust method for non-rigid point set registration, using the Gaussian mixture model and accommodating non-rigid transformations. The posterior probabilities of the mixture model are determined through the proposed integrated feature divergence.

The method involves an alternating two-step framework, comprising correspondence estimation and subsequent transformation updating. For correspondence estimation, integrated feature divergences including both global and local features, are coupled with deterministic annealing to address the non-convexity problem of registration. For transformation updating, the expectation-maximization iteration scheme is introduced to iteratively refine correspondence and transformation estimation until convergence.

The experiments confirm that the proposed registration approach exhibits remarkable robustness on deformation, noise, outliers and occlusion for both 2D and 3D point clouds. Furthermore, the proposed method outperforms existing analogous algorithms in terms of time complexity. Application of stabilizing and securing intermodal containers loaded on ships is performed. The results demonstrate that the proposed registration framework exhibits excellent adaptability for real-scan point clouds, and achieves comparatively superior alignments in a shorter time.

The integrated feature divergence, involving both global and local information of points, is proven to be an effective indicator for measuring the reliability of point correspondences. This inclusion prevents premature convergence, resulting in more robust registration results for our proposed method. Simultaneously, the total operating time is reduced due to a lower number of iterations.

The purpose of this study is to introduce the concept of Ecopoiesis. Introduction of the concept of ecopiesis, which explains the linkage strategy between relational and energetic resources (sustainability) that enable the relational viability of living things.

The design is based on the use of the relationship between abduction, epigenesis, cismogenesis and morphogenesis.

This paper concludes with a counter-intuitive proposal to the dominant lines of thought, distinguishing: Complexity as vital interweaving, Aristotelian Oikos as communitarian distinction, Organism-Entorno from relational theory (beyond the interactional), in a cybernetic referential framework that allows tending to sustain and sustain life from Ecopoiesis.

According to a relational theory of cognition, the sources of knowledge go beyond genetic and environmental factors. In the sphere of explanation, this study moves from “how much,” “why” and “how” to “why not.” Thus, the processes involved in understanding the living are forced to move from the logic of parts to the logic of patterns, which implies developing an explanation that considers the unique characteristics of development, encompassing the diverse origins of the organization of the living.

This study proposes a collaborative federated learning (CFL) framework to address personal data transmission and retention issues for artificial intelligence (AI)-enabled video surveillance in public spaces.

This study examines specific challenges for long-term people monitoring in public spaces and defines AI-enabled video surveillance requirements. Based on the requirements, this study proposes a CFL framework to gradually adapt AI models’ knowledge while reducing personal data transmission and retention. The framework uses three different federated learning strategies to rapidly learn from different new data sources while minimizing personal data transmission and retention to a central machine.

The findings confirm that the proposed CFL framework can help minimize the use of personal data without compromising the AI model's performance. The gradual learning strategies help develop AI-enabled video surveillance that continuously adapts for long-term deployment in public spaces.

This study makes two specific contributions to advance the development of AI-enabled video surveillance in public spaces. First, it examines specific challenges for long-term people monitoring in public spaces and defines AI-enabled video surveillance requirements. Second, it proposes a CFL framework to minimize data transmission and retention for AI-enabled video surveillance. The study provides comprehensive experimental results to evaluate the effectiveness of the proposed framework in the context of facial expression recognition (FER) which involves large-scale datasets.

This paper explores the orchestration of digital innovation in Industry 4.0 organisations.

The study applies the activity theory to explorative multiple case studies. Observations of innovation activities in five business cases take place at two large international organisations.

The results underline five logics of action that drive digital innovation: (1) digital transformation, (2) technology translation, (3) catalyst agents, (4) digital thread and (5) empowerment. Further, the case study organisations highlight the importance of developing a sustainable culture capable of continuously adopting new technologies, processes and infrastructure that will allow the management of digital innovations.

The study empirically shows the motivations and challenges in orchestrating digital innovation in Industry 4.0 organisations.

This study investigates how individuals' self-awareness (specifically, private and public self-awareness) and environment-awareness (perceived expertise, similarity and familiarity) shape herd behavior, encompassing discounting one’s information and imitating others. Drawing from latent state-trait theory, this research aims to discern the impact of these factors on purchase intention and behavior.

Longitudinal data from 231 users in Xiaohongshu, China’s leading social commerce platform, were collected to test the proposed model and hypotheses.

The findings from this study show that private self-awareness negatively influences discounting one’s own information and imitating others. Public self-awareness positively affects imitating others, while it does not affect discounting one’s own information. Perceived expertise diminishes discounting one’s own information but does not significantly affect imitating others. Perceived similarity and perceived familiarity are positively related to discounting one’s own information and imitating others. The results confirm different interaction effects between self-awareness and environment-awareness on herd behavior.

First, this contributes back to the latent state-trait theory by expanding the applicability of this theory to explain the phenomenon of herd behavior. Second, this study takes an important step toward theoretical advancement in the extant literature by qualifying that both self- and environment-awareness should be considered to trigger additional effects on herd behavior. Third, this study provides a more enlightened understanding of herd behavior by highlighting the significance of considering the interplay between self- and environment-awareness on herd behavior. Finally, this study also empirically confirms the validity of classifying self-awareness into private and public aspects.