Search results

The paper aims to analyse and compare how UK and Singapore deal with compensation with respect to regulation of land (short of a physical taking). The purpose is to determine whether the non-compensation in each jurisdiction is justified.

A comparative method using case law, statutes and secondary material across both jurisdictions (as well as some US case law) is adopted.

Both the UK and Singapore do not provide compensation when land is affected by regulation, so long as a physical taking has not occurred. Partly because of the abolition of development rights in the UK since 1947, this position may be justified. Conversely, Singapore’s Master Plan seeks a great deal of public reliance and advertises development potential, and non-compensation is not defensible.

There is very limited analysis on regulatory effects of land in the UK, and virtually none in Singapore. This would also be the first attempt to compare this aspect of the UK and Singapore’s planning regime.

Since China initiated its “go global” policy that promotes its overseas investment, China’s Outward Foreign Direct Investment (OFDI) has increased almost twenty times during the last 10 years, reaching $55.9 billion in 2008. The issue of internationalization of Chinese OFDI has attracted increasing attention of researchers from a business perspective. This article systematically reviews the previous studies on overseas investments by Chinese MNEs and discusses the characteristics of Chinese internationalization behavior at both firm level and country level. The internationalization of Chinese companies cannot be understood as a simple game of “catch up” with established MNEs, and more firm‐level empirical studies should be carried out on how these characteristics influence firms’ strategic decisions.

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.