Search results

Typical feature-matching algorithms use only unary constraints on appearances to build correspondences where little structure information is used. Ignoring structure information makes them sensitive to various environmental perturbations. The purpose of this paper is to propose a novel graph-based method that aims to improve matching accuracy by fully exploiting the structure information.

Instead of viewing a frame as a simple collection of keypoints, the proposed approach organizes a frame as a graph by treating each keypoint as a vertex, where structure information is integrated in edges between vertices. Subsequently, the matching process of finding keypoint correspondence is formulated in a graph matching manner.

The authors compare it with several state-of-the-art visual simultaneous localization and mapping algorithms on three datasets. Experimental results reveal that the ORB-G algorithm provides more accurate and robust trajectories in general.

Instead of viewing a frame as a simple collection of keypoints, the proposed approach organizes a frame as a graph by treating each keypoint as a vertex, where structure information is integrated in edges between vertices. Subsequently, the matching process of finding keypoint correspondence is formulated in a graph matching manner.

Conventional learning-based visual odometry (VO) systems usually use convolutional neural networks (CNN) to extract features, where some important context-related and attention-holding global features might be ignored. Without essential global features, VO system will be sensitive to various environmental perturbations. The purpose of this paper is to design a novel learning-based framework that aims to improve accuracy of learning-based VO without decreasing the generalization ability.

Instead of CNN, a context-gated convolution is adopted to build an end-to-end learning framework, which enables convolutional layers that dynamically capture representative local patterns and composes local features of interest under the guidance of global context. In addition, an attention mechanism module is introduced to further improve learning ability and enhance robustness and generalization ability of the VO system.

The proposed system is evaluated on the public data set KITTI and the self-collected data sets of our college building, where it shows competitive performance compared with some classical and state-of-the-art learning-based methods. Quantitative experimental results on the public data set KITTI show that compared with CNN-based VO methods, the average translational error and rotational error of all the test sequences are reduced by 45.63% and 37.22%, respectively.

The main contribution of this paper is that an end-to-end deep context gate convolutional VO system based on lightweight attention mechanism is proposed, which effectively improves the accuracy compared with other learning-based methods.

Although trust is generally taken as a fundamental factor in influencing relational behavior in contractor–subcontractor collaboration, the determination of an optimal level of trust is still lacking. Trust with an optimal tipping point that matches dependence best is considered the optimal trust to improve relational behavior between general contractors and subcontractors. To fill the knowledge gap, this study explores how combinations of trust and dependence trigger relational behavior between general contractors and subcontractors through a configurational approach.

Questionnaires were administered to 228 middle management and technical staff members of the general contractor. The data were analyzed using fuzzy-set qualitative comparative analysis (fsQCA), and the inductive analytic method allowed researchers to explore configurations of different dimensions and levels of dependence and trust.

Necessity analysis results indicated that neither dependence nor trust was a necessary condition for facilitating relational behavior. Through sufficiency analysis, four configurations of optimal trust matched with dependence were identified in contractor–subcontractor collaboration. Even if contractors rely only on subcontractors for resources, the optimal trust between contractors and subcontractors should include both institution- and cognition-based trust. In the event that contractor–subcontractor collaboration involves relational dependence, both affect- and cognition-based trust are necessary for the optimal trust.

This study enhances existing research by delving deeper into a nuanced understanding of optimal trust in dependence scenarios, and enriches project governance theory by uncovering the internal transmission of relational governance. Practically, this study offers general contractors guidance on how to establish optimal trust strategies based on the dual dependence level with subcontractors, which can facilitate subcontractors' relational behavior, and ultimately improve contractor–subcontractor collaboration performance.

The purpose of this study is to restore the spatial distribution of overseas remittance businesses in Shantou during the 1940s. It explores various socioeconomic factors that influenced the concentration of local remittance business investment in real estate. By reconstructing the spatial distribution of remittance business activities in Shantou, this study hopes to lay a foundation for further analysis of the business strategies of Chaoshan merchants.

This research draws on information from the published Swatow Guide, archival sources and cadastral maps to identify the location of remittance enterprises and the native place and overseas networks of property owners.

This study reveals that the spatial distribution of the remittance enterprises was determined by the native place origins of local property owners, and that the inflow of overseas Chinese capital contributed to real estate development in Shantou.

Despite the limited access to Chinese official archives, this paper manages to identify several building blocks and neighbors in Shantou for spatial analysis.

This study is the first attempt to use the geographical information system (GIS) method in Chinese urban history research and hopes to establish a larger historical database of Shantou as a sample for comparison.

This investigation advances the spatial study of urban history and overseas Chinese remittances in the maritime society of South China.

The purpose of this study is to investigate whether the corporate environmental, social and governance (ESG) performance of enterprise is influenced by the enterprise digital transformation. In addition, this study explains how enterprise digital transformation affects ESG performance.

The sample covers 4,646 nonfinancial companies listed on China’s A-share market from 2009 to 2021. The study adopts the fixed-effects multiple linear regression to perform the data analysis.

The study finds that enterprise digital transformation has a significant inverted U-shaped impact on ESG performance. Moderate digital transformation can improve enterprise ESG performance, whereas excessive digital transformation will bring new organizational conflicts and increase enterprise costs, which is detrimental to ESG performance. This inverted U-shaped effect is more pronounced in industrial cities, manufacturing industries and enterprises with less financing constraints and executives with financial backgrounds. Enterprise digital transformation mainly affects ESG performance by affecting the level of internal information communication and disclosure, the level of internal control and the principal-agent cost.

The government should take multiple measures to encourage enterprises to choose appropriate digital transformation based on their own production behaviors and development strategies, encourage them to innovate and upgrade their organizational management and development models in conjunction with digital transformation and guide them to use digital technology to improve ESG performance.

This study shows that irrational digital transformation cannot effectively improve the ESG performance of enterprises and promote the sustainable development of the country. Enterprises should carry out reasonable digital transformation according to their own development needs and finally improve the green and sustainable development ability of enterprises and promote the sustainable development of society.

This study examines the relationship between enterprise digital transformation and ESG performance. Different from the linear relationship between the two in previous major studies, this study proves the inverse U-shaped relationship between enterprise digital transformation and ESG performance through mathematical theoretical model derivation and empirical test. This study also explores in detail how corporate digital transformation affects ESG performance, as well as discusses heterogeneity at the city, industry and firm levels. It is proposed that enterprises should take into account their own characteristics and carry out reasonable digital transformation according to their development needs.