Spatial Perception Research Articles

AUTOTELICITY AS A CAPACITY OF ARCHITECTURAL CONCEPT: ARCHITECTURE OF CREATIVITY IN CHILDREN'S SPACES

<p>This paper explores the intricate relationship between architecture and creativity, with a particular emphasis on the concept of autotelicity. Autotelicity, defined as a state where an individual fully immerses themselves in an activity within a spatial context, is examined as a crucial element in the creative process and appreciation of architecture. The investigation revolves around how autotelicity can form the basis for architectural concepts and how architecture, in turn, can encourage user creativity through this lens. Through case studies of children's spaces, this research aims to illuminate these concepts and provide fresh insights into how architecture influences our spatial perception while fostering creative thought and behavior. The emphasis on autotelicity underscores architecture's potential to inspire creativity and enjoyment in everyday spatial use.</p>

A study on commuters’ public transportation mode choice behavior in river valley-type cities considering terrain spatial perception: evidence from Lanzhou, China

Existing research rarely examines the subjective and objective built environment of river valley-type cities in relation to travel mode choice, particularly overlooking the heterogeneity among travelers in these cities. In this paper, based on questionnaire survey data and built environment data, terrain spatial perception (TSP) is introduced to expand the theory of planned behavior (TPB), and a Structural Equation Model (SEM) is established. Factor analysis and path analysis are conducted using SPSS and AMOS to estimate latent variables. An integrated model of SEM and random parameter Logit model (RPLM), which can not only analyze the psychological perception factors of commuters in river valley-type cities but also consider the heterogeneity of psychological perception, was constructed to analyze the impact of personal attributes, objective built environment factors, and psychological latent variables on the commuting mode choice behavior of public transport users in river valley-type cities. The results indicate that the five observation indicators corresponding to the proposed terrain spatial perception latent variables can better explain the terrain spatial perception of commuters in river valley-type cities. Different from plain cities, the subjective and objective built environment of river valley-type cities notably influence the travel behavior of commuters. Moreover, the parameters of terrain spatial perception follow a normal distribution, indicating that the sensitivity of different commuters to the terrain spatial perception of river valley-type cities is heterogeneous. The results of our study can provide a reference for alleviating traffic issues in valley cities.

Analysis of one-year outcomes after cochlear implantation in adults with single-sided deafness

Objectives: To investigate the outcomes of cochlear implantation in Mandarin-speaking cochlear implant (CI) users with single-sided deafness (SSD). Methods: This study was a single-center prospective cohort study. Eleven Mandarin-speaking adult SSD patients who underwent CI implantation at Capital Medical University Beijing Tongren Hospital from August 2020 to October 2021 were recruited, including 6 males and 5 females, with the age ranging from 24 to 50 years old. In a sound field with 7 loudspeakers distributed at 180°, we measured root-mean-square error(RMSE)in SSD patients at the preoperative, 1-month, 3-month, 6-month, and 12-month after switch-on to assess the improvement of sound source localization. The Mandarin Speech Perception (MSP) was used in the sound field to test the speech reception threshold (SRT) of SSD patients under different signal-to-noise locations in a steady-state noise under conditions of CI off and CI on, to reflect the head shadow effect(SSSDNNH), binaural summation effect(S0N0) and squelch effect(S0NSSD). The Tinnitus Handicap Inventory (THI) and the Visual Analogue Scale (VAS) were used to assess changes in tinnitus severity and tinnitus loudness in SSD patients at each time point. The Speech, Spatial and Qualities of Hearing Scale(SSQ) and the Nijmegen Cochlear Implantation Scale (NCIQ) were used to assess the subjective benefits of spatial speech perception and quality of life in SSD patients after cochlear implantation. SPSS 19.0 software was used for statistical analysis. Results: SSD patients showed a significant improvement in the poorer ear in hearing thresholds with CI-on compared with CI-off; The ability to localize the sound source was significantly improved, with statistically significant differences in RMSE at each follow-up time compared with the preoperative period (P<0.05). In the SSSDNNH condition, which reflects the head shadow effect, the SRT in binaural hearing was significantly improved by 6.5 dB compared with unaided condition, and the difference was statistically significant (t=6.25, P=0.001). However, there was no significant improvement in SRT between the binaural hearing condition and unaided conditions in the S0N0 and S0NSSD conditions (P>0.05). The total score of THI and three dimensions were significant decreased (P<0.05). Tinnitus VAS scores were significantly lower in binaural hearing compared to the unaided condition (P<0.001). The total score of SSQ, and the scores of speech and spatial dimensions were significant improved in binaural hearing compared to the unaided condition (P<0.001). There was no statistical difference in NCIQ questionnaire scores between preoperative and postoperative (P>0.05), and only the self-efficacy subscore showed a significant increase(Z=-2.497,P=0.013). Conclusion: CI could help Mandarin-speaking SSD patients restore binaural hearing to some extent, improve sound localization and speech recognition in noise. In addition, CI in SSD patients could suppress tinnitus, reduce the loudness of tinnitus, and improve subjective perceptions of spatial hearing and quality of life.

Enhancing target detection accuracy through cross-modal spatial perception and dual-modality fusion

The disparity between human and machine perception of spatial information presents a challenge for machines to accurately sense their surroundings and improve target detection performance. Cross-modal data fusion emerges as a potential solution to enhance the perceptual capabilities of systems. This article introduces a novel spatial perception method that integrates dual-modality feature fusion and coupled attention mechanisms to validate the improvement in detection performance through cross-modal information fusion. The proposed approach incorporates cross-modal feature extraction through a multi-scale feature extraction structure employing a dual-flow architecture. Additionally, a transformer is integrated for feature fusion, while the information perception of the detection system is optimized through the utilization of a linear combination of loss functions. Experimental results demonstrate the superiority of our algorithm over single-modality target detection using visible images, exhibiting an average accuracy improvement of 30.4%. Furthermore, our algorithm outperforms single-modality infrared image detection by 3.0% and comparative multimodal target detection algorithms by 3.5%. These results validate the effectiveness of our proposed algorithm in fusing dual-band features, significantly enhancing target detection accuracy. The adaptability and robustness of our approach are showcased through these results.

Evaluating gender-based effectiveness within multi-attribute task battery studies

The prevalent notion that females outperform males in multitasking is challenged by scrutinizing the cognitive abilities of each gender across various demand scenarios (Buser &amp; Peter, 2012; Lui et al., 2020; Szameitat et al., 2015). Investigating cognitive performance at two levels of workload defined as high- and low-demand situations, this research delves into task-specific variations, cognitive strategies, and adaptability to multitasking challenges. Sixty participants in total, with thirty participants in each gender group, engaged in tasks encompassing resource management, tracking, monitoring, and communication using the U.S. Army Aeromedical Research Laboratory's Multi- Attribute Task Battery simulation software. Statistical variance analysis demonstrated notable task discrepancies across different difficulty levels (DL). However, the gender-based analysis indicated non-significant differences in given tasks, except communication tasks. Findings indicate that men tend to perform well in tasks requiring communication and target tracking, benefiting from auditory cues, while women excel in fuel management, displaying strong decision-making abilities. In assessments of visual and spatial perception, females outperformed males at low difficulty levels, whereas males showed greater proficiency at high difficulty levels. This paper advocates for a balanced understanding and approach of cognitive abilities in diverse scenarios.

РОМАНТИЗМ И КОНСТРУКТИВИЗМ В ПОЛЬСКОМ ТЕАТРЕ 1920-1930-Х ГОДОВ

In the 1920-1930s, Polish theatre was experiencing a turn to new comprehension of the practice of stage space organization – the art of scenography was developing. It replaced the practice of creating theatrical decorations for garnishing the play and constituting the background for the action. The new scenography declared itself to be an art of the stage space composing. The development of new theatrical practice was influenced by constructivist spatial and rhythmical perception of environment, which Polish artists were improving in close cooperation with avant-garde architects. A similar approach to the stage design is common for the whole European theatre in the first decades of XX century, however the distinction of theatrical avant-garde in Poland came out as a special reference of the culture to its past and in particular to the romanticism heritage. Polish theatrical avant-garde, which came at restoration of independence time, was developing on overlap of the national myth and new unexplored freedom. Artists of the first half of XX century managed to disengage from the romantic myth only in some measure, but instead of resetting the national heroes and their heritage from the ship of modernity they turned to search for the new forms to embody that matter. The article reviews Polish theatre of the 1920-1930-s in the context of Polish romantic tradition on the one hand and constructivism, promoted by new avant-garde artist collectives – on the other. As an example of integration of two movements, a brief description of the unique project of Simultaneous theatre, created by stage designer Andrzej Pronaszko and architect Szymon Syrkus is given.

Diversity-Learning Block: Conquer Feature Homogenization of Multibranch.

Visual Geometry Group (VGG)-style ConvNet is an neural-network process units (NPU)-friendly network; however, the accuracy of this architecture cannot keep up with other well-designed network structures. Although some reparameterization methods are proposed to remedy this weakness, their performance suffers from the homogenization issue of parallel branches, and the preset shape of convolution kernels also influences spatial perception. To address this problem, we propose a diversity-learning (DL) block to build the DLNet, which could adaptively learn various features to enrich the feature space. To balance floating point of operations (FLOPs) and accuracy, groupwise operation is introduced and finally, a lightweight DL ConvNet DLGNet is obtained. Extensive evaluations have been conducted on different computer vision tasks, e.g., image classification [Canadian Institute For Advanced Research (CIFAR) and ImageNet], object detection [PASCAL visual object classes (VOC) and Microsoft Common Objects in Context (MS COCO)], and semantic segmentation (Cityscapes). The experimental results show that our proposed DLGNet can achieve comparable performance with the state-of-the-art networks while the speed is 183% faster than GhostNet and even over 600% faster than MobileNetV3 with similar accuracy when running on NPU.

Enhanced-YOLOv8: A new small target detection model

Small target detection is a very difficult task and remains one of the most challenging problems in computer vision due to variations in object shape, appearance and position, as well as the effects of lighting and occlusion during imaging. To improve the accuracy of the small target detection results, we propose a new small target detection model, Enhanced-YOLOv8. Firstly, a small target detection level (STDL) is added to the original effective feature layer of YOLOv8, which not only provides richer semantic information but also get more accurate target localization and bounding box accuracy. When detecting small targets, the detection accuracy can be improved by more detailed information. Then, the fusion convolutional block attention module (FCBAM) is proposed by introducing the position attention module (PAM) based on the traditional CBAM. FCBAM not only can adaptively select and fuse the most representative features, but also can better capture the image important features at different positions in the image and enhance spatial detail perception. Finally, semantic fusion network (SFN) is proposed on the basis of residual network, which introduces semantic information of high-layer feature into low-layer feature. It can adaptively guide the fusion of high-layer feature and low-layer feature to reduce the loss of feature information. After experimental verification, the Enhanced-YOLOv8 proposed improves the accuracy of the detection results.

Indoor navigation map design based on spatial complexity

ABSTRACT In the face of the complex interior structure of a building, indoor navigation is an important tool for people to arrive at their destination in large public buildings. In this research, we first analyze the complexity of indoor space, which mainly includes spatial entities with rich semantics, but corridors lacking in semantics, poor visibility, and multi-dimensional floors. Then, we designed various indoor navigation maps based on spatial complexity and conducted a comparative analysis between the existing maps and the ones we designed. Participants were recruited to perform three navigation tasks in the study area on-site. The results demonstrate the advantages of using a hierarchical landmark representation for improved spatial knowledge acquisition and the effectiveness of scene-enhanced directional guidance in areas with poor visibility. It is worth noting that the introduction of three-dimensional maps was found to have a positive impact on users’ spatial perception and their ability to sketch routes accurately. In summary, this evidence confirms that maps based on spatial complexity not only enhance navigation efficiency but also have the potential to improve the overall indoor navigation experience.

Multi-path parallel enhancement of low-light images based on multiscale spatially aware Retinex decomposition

To resolve the issues of color distortion and detail blurring, this study proposes a multi-path parallel enhancement of low-light images (LLIs) based on multiscale spatially aware Retinex decomposition (MUSRetinex-Net). MUSRetinex-Net consists of three subnetworks. Firstly, MUSRetinex-Net inputs LLIs into an illumination level estimation image guided multiscale spatial perception decomposition network (MSPD-Net) which completes Retinex decomposition to retain richer color information and detail structure; then reflectance maps from LLIs are inputted into a three-pass parallel reflectance maps recovery network (TPR-Net) for more desirable noise removal and color recovery; illumination maps from LLIs are inputted into a multi-branch attention illumination maps adjustment network (MAA-Net) for adjustment to obtain a more accurate estimation of illumination levels. In this study, six publicly accessible datasets are utilized for experiments, two referenced evaluation metrics, and four non-referenced evaluation metrics are utilized for assessment and analysis. The experimental results show that the low-light image enhancement (LLIE) effects obtained by this study method are more consistent with the recognition level of the human eye and the structure of the natural environment. The method achieves the optimal referenced evaluation metrics score and the most comprehensive non-referenced evaluation metrics score on all six public datasets, with extremely high enhancement performance and generalization ability.

A study on optimizing public spaces in rural heritage tourist sites from spatial perception: The case of Jinggang historic town

The development of cultural and rural tourism has fostered the expansion of heritage sites while concurrently presenting challenges such as resource wastage and cultural degradation. Optimizing spatial use requires a comprehensive understanding of the dynamics inherent to these sites. An emerging trend involves the integration of human perception technology into spatial perception research. Thus, to examine the correlation between alterations in physical and visual environments, tourists&amp;rsquo; perceived changes, and primary influencing elements in rural heritage tourism sites, Jinggang &amp;ndash; a historic and cultural town in Changsha, Hunan province, China &amp;ndash; was selected as the research site. A comprehensive interdisciplinary analysis was conducted by integrating physical environment data from public spaces. These data include air temperature (Ta), relative humidity (RH), wind speed (WS), equivalent sound pressure level (ESPL), visual environment data (including green view index [GVI], sky view index [SVI], road and pavement view index (R&amp;PVI), and construction view index [CVI]), along with biosensing data (level of stress [LS]) collected from tourists. The findings indicate the following: (i) the LS of the public space in Jinggang falls within the normal range but remains generally high, with the humanistic landscape, dominated by traditional architecture, scoring higher than the natural landscape; (ii) regarding spatial perception, the quantity of physical environment elements influencing LS surpasses that of visual environment elements, and the most relevant indicators, including ESPL, R&amp;PVI, and CVI, all contribute to tourists&amp;rsquo; LS; and (iii) ESPL significantly influences LS within the environment domain, where higher environmental noise decibels correspond to tourists&amp;rsquo; increased LS. The study concludes that historic towns possess distinct characteristics as types of rural heritage tourism sites. This study provides valuable insights into enhancing spatial environmental quality and promoting an appealing and sustainable rural setting by implementing public space optimization strategies in rural heritage sites.

Beyond built environment: Unveiling the interplay of streetscape perceptions and cycling behavior

As an important means of shared micro-mobility, shared bicycles have become a crucial component of urban transportation in China. The impact of the built environment on bicycling has been widely acknowledged. However, can streetscape perceptions influence bicycle-sharing volume (BSV) and supplement the built environment? We first obtained millions of pieces of shared-cycling data from the Shenzhen Open Data Platform and carried out geographical quantification of BSV. As for streetscape, we improved the classification of subjective streetscape perception based on street view images using the k-means clustering algorithm and conducted predictions using XGBoost. Through the application of different regression models, we unveiled the nonlinear spatial interdependencies between BSV and streetscape perceptions as a complement to the built environment. Our findings indicate that greenery, vivid street-front facades, and diverse street facilities can promote BSV. Targeted strategies are proposed for different districts. For instance, urban planners can provide incentives for high-income groups in central urban areas to adopt active travel, and increase the supply of shared bicycles in suburban areas with high building density, particularly in industrial urban villages. As a supplement to the long-term planning recommendations derived from the macro-built environment analysis, an in-depth spatial perception quantitative assessment proffers a human-centric, flexible blueprint for urban street design.

Towards a Communication Ecology in the Life of Rural Senior Citizens: How Rural Public Spaces Influence Community Engagement

The dilemma of weak participation and non-participation of rural communities is a universal topic of global development. The rural public space is an important field for local residents to interact, communicate, and engage with each other, and is an important place for the sustainable development of rural areas. However, previous studies have neglected to understand the intrinsic connection between rural public space and community participation from the perspective of community communication ecology. Based on the concept of age-friendly communities, this study’s fieldwork in rural Shanghai, China, using the methodology of grounded theory, found that physical, social, and psychological factors all have an impact on community engagement among rural residents. Specifically, environmental quality, facility support, community networks, social participation, call to action, place attachment, spatial perception, and self-transformation are identified as the core elements that significantly influence community engagement among rural residents. This study further reveals that the multiple factors influencing community engagement among rural residents are complex and interdependent rather than operating independently. Spatial support, communicative triggers, and symbolic identification, respectively, operate at the technological level (physical–social factors), social level (social–psychological factors), and discursive level (psychological–physical factors) to promote community engagement among older adults in Chinese rural areas. Through this research, we hope to further the realization of rural civic engagement and the sustainability of local communities and to provide scholarly insights into the promotion of more equitable community life.

ВЛАСТИВОСТІ АРХІТЕКТУРНО-МІСТОБУДІВНОГО ТВОРЕННЯ СЕРЕДМІСТЬ У ОДНОЙМЕННИХ МІСТАХ ДЕРАЖНЕ І ДЕРАЖНЯ

The properties of the downtown architectural and town-planning creation in the historical cities of Derazhne and Derazhnya are considered. It has been found that in Derazhne the castle was located in the southern part of today's village, on a wedge ledge formed at the place where the stream flows into the Horyn River. The downtown stood northeast of the castle. Its territory had features of a regular rectangle, and the gates were located on the western and northern sides. Unfortunately, today it is difficult to determine the location of the wooden Roman Catholic church in the downtown, instead, the brick church of the beginning of the 19th century was built on the territory of the castle. It was established that the castle in Deraznya was rectangular in plan with two corner roundels located in the southern and northern corners of the western side. The indistinct fragments of the downtown ramparts preserved to this day, the configuration of the streets and the graphic image of the urban defense complex of Deraznya on the map of the mid-17th century by the military engineer Guiyaume Levasseur de Beauplan indicate that the northern and eastern parts of the fortifications of the city center had bastion-like protrusions, and the northeastern corner had a special fortification purpose . Rynok Square was located in the center of the downtown. A wooden church stood in its northeastern corner, and a brick Roman Catholic church in the southeastern corner. Relying on the determined properties of architectural and town-planning creation in the downtown of Derazhnye and Derazhnya settlements, we can record their equivalent historical and cultural value. The development vector of their cities was identical, but the architectural image was individual. The spatial composition of the connection between the castle and the city center, the abundance of temples and the fortification system differed. Instead, the street network in Derazne and Derazne was ordered, calculated and synchronized with two entrance gates. The current state of the downtown is unsatisfactory. In Derazhnya, Khmelnytskyi region, the street network, quarters with several houses of the 19th century, and castle ramparts have been partially preserved. The downtown of Derazne, Rivne Oblast, has preserved parts of historic streets, completely lost quarters, and faint traces of the castle. In both settlements, uncontrolled green spaces hinder the spatial perception of the historical center. Therefore, recommendations have been formulated for the return of service functions to the downtowns, their unification with the rest of the settlement territories, saturating them with new functions, and the reproduction of the traditional historical and urban landscape.

Exploring the Role of Video Playback Visual Cues in Object Retrieval Tasks.

Searching for objects is a common task in daily life and work. For augmented reality (AR) devices without spatial perception systems, the image of the object's last appearance serves as a common search assistance. Compared to using only images as visual cues, videos capturing the process of object placement can provide procedural guidance, potentially enhancing users' search efficiency. However, complete video playback capturing the entire object placement process as visual cues can be excessively lengthy, requiring users to invest significant viewing time. To explore whether segmented or accelerated video playback can still assist users in object retrieval tasks effectively, we conducted a user study. The results indicated that when video playback is covering the first appearance of the object's destination to the object's final appearance (referred to as the destination appearance, DA) and playing at normal speed, search time and cognitive load were significantly reduced. Subsequently, we designed a second user study to evaluate the performance of video playback compared to image cues in object retrieval tasks. The results showed that combining the DA playback starting point with images of the object's last appearance further reduced search time and cognitive load.

Pushing the Limits of Acoustic Spatial Perception via Incident Angle Encoding

With the growing popularity of smart speakers, numerous novel acoustic sensing applications have been proposed for low-frequency human speech and high-frequency inaudible sounds. Spatial information plays a crucial role in these acoustic applications, enabling various location-based services. However, typically commercial microphone arrays face limitations in spatial perception of inaudible sounds due to their sparse array geometries optimized for low-frequency speech. In this paper, we introduce MetaAng, a system designed to augment microphone arrays by enabling wideband spatial perception across both speech signals and inaudible sounds by leveraging the spatial encoding capabilities of acoustic metasurfaces. Our design is grounded in the fact that, while sensitive to high-frequency signals, acoustic metasurfaces are almost non-responsive to low-frequency speech due to significant wavelength discrepancy. This observation allows us to integrate acoustic metasurfaces with sparse array geometry, simultaneously enhancing the spatial perception of high-frequency and low-frequency acoustic signals. To achieve this, we first utilize acoustic metasurfaces and a configuration optimization algorithm to encode the unique features for each incident angle. Then, we propose an unrolling soft thresholding network that employs neural-enhanced priors and compressive sensing for high-accuracy, high-resolution multi-source angle estimation. We implement a prototype, and experimental results demonstrate that MetaAng maintains robustness across various scenarios, facilitating multiple applications, including localization and tracking.

Classification of Neurocognition in Japanese Patients with Schizophrenia: A Cluster Analysis

Objectives. Cognitive functions in almost all domains are lower in patients with schizophrenia than those in healthy controls, with the severity of impairment differing between domains. Treatments are being developed to improve cognitive impairment in patients with schizophrenia. However, the pattern of cognitive impairment must be clarified to facilitate treatment. Therefore, this study aimed to classify the patterns of cognitive impairment in individuals and provide treatment suggestions. Methods. Patients with schizophrenia were recruited from two psychiatric hospitals in Japan. Demographic and psychopathological symptoms were assessed using the Positive and Negative Symptoms Scale for Schizophrenia and neurocognitive functions, using the CogHealth battery. The following domains were assessed: processing speed, visual attention, working memory, visual learning, and spatial attention. The scores were standardised and assigned as the same-aged average score. Hierarchical cluster analysis using Ward’s method was performed based on CogHealth scores. Subsequently, one-way analysis of variance (ANOVA) and Tukey’s multiple comparisons were performed to compare the variables in each cluster. Results. In total, 133 participants were classified into four clusters: Cluster 1 (n = 16), with severe cognitive impairment and psychiatric symptoms and the longest stay; Cluster 2 (n = 44), with moderate cognitive impairment and psychiatric symptoms; Cluster 3 (n = 42), with preserved cognitive function, except for spatial perception, and mild psychiatric symptoms; and Cluster 4 (n = 31), with only memory and spatial perception impairment and mild psychiatric symptoms. Implications. The clusters indicate that impairment may occur in all or selective domains. Selective domain impairments may be in spatial perception or in spatial perception and memory. Therefore, it is recommended that treatments for cognitive dysfunction are developed into four subsets considering an individual’s cognitive features.

The perceptual effect of forms used on walls and furnishings in cafés

PurposeThis study researched the effects on the perceptual evaluations of participants for the physical environmental factors of cafés using curvilinear, rectilinear, and mixed forms to provide for the spatial states of belonging, to increase the pleasures, and to extend the periods of remaining in the space.Design/methodology/approachThis study used virtual reality (VR) technology to model the cafés designed by using curvilinear, rectilinear and mixed forms as an experimental environment. After experiencing the virtual images of the experimental spaces, participants filled out a “spatial perception” questionnaire. Perceptual evaluations of 415 participants regarding the environmental factors of cafés designed using curvilinear, linear and mixed forms were analyzed in a computer environment.FindingsAccording to the analysis of the questionnaire data, the cafés using mixed forms were perceived as more positive for the factors of appeal (inviting, restful, warm and sincere), planning (well-planned) and space freedom (roomy, uncluttered, uncrowded, large, wide and free space) compared with the cafés using curvilinear and rectilinear forms. Furthermore, the cafés using curvilinear forms were perceived as more positive for all elements compared with the cafés using rectilinear forms. However, there was a more negative approach in the perceptual evaluations of participants connected to increases in level of education.Originality/valueThe research results clearly demonstrated that the different interior and furniture forms frequently encountered in cafés cause significant effects on users’ perceptual evaluations.

Filmic space

The projects collected here under the banner of Filmic Space are seen as an extension of the ideas brought to light in the publication Interior Cities. Interior Cities represents the outcomes of the last seven years of the program grouped within specific conceptual concerns. The book chronicles ideas of spatial perception, sensing space, elemental space, contextual space, cross cultural space, space of the imagination, incisions into the metropolis and virtual space.

An attention mechanism module with spatial perception and channel information interaction

In the field of deep learning, the attention mechanism, as a technology that mimics human perception and attention processes, has made remarkable achievements. The current methods combine a channel attention mechanism and a spatial attention mechanism in a parallel or cascaded manner to enhance the model representational competence, but they do not fully consider the interaction between spatial and channel information. This paper proposes a method in which a space embedded channel module and a channel embedded space module are cascaded to enhance the model’s representational competence. First, in the space embedded channel module, to enhance the representational competence of the region of interest in different spatial dimensions, the input tensor is split into horizontal and vertical branches according to spatial dimensions to alleviate the loss of position information when performing 2D pooling. To smoothly process the features and highlight the local features, four branches are obtained through global maximum and average pooling, and the features are aggregated by different pooling methods to obtain two feature tensors with different pooling methods. To enable the output horizontal and vertical feature tensors to focus on different pooling features simultaneously, the two feature tensors are segmented and dimensionally transposed according to spatial dimensions, and the features are later aggregated along the spatial direction. Then, in the channel embedded space module, for the problem of no cross-channel connection between groups in grouped convolution and for which the parameters are large, this paper uses adaptive grouped banded matrices. Based on the banded matrices utilizing the mapping relationship that exists between the number of channels and the size of the convolution kernels, the convolution kernel size is adaptively computed to achieve adaptive cross-channel interaction, enhancing the correlation between the channel dimensions while ensuring that the spatial dimensions remain unchanged. Finally, the output horizontal and vertical weights are used as attention weights. In the experiment, the attention mechanism module proposed in this paper is embedded into the MobileNetV2 and ResNet networks at different depths, and extensive experiments are conducted on the CIFAR-10, CIFAR-100 and STL-10 datasets. The results show that the method in this paper captures and utilizes the features of the input data more effectively than the other methods, significantly improving the classification accuracy. Despite the introduction of an additional computational burden (0.5 M), however, the overall performance of the model still achieves the best results when the computational overhead is comprehensively considered.