Dynamic hand gesture recognition is a crucial yet challenging task in computer vision. The key of this task lies in an effective extraction of discriminative spatial and temporal features to model the evolutions of different gestures. In this paper, we propose an end-to-end Spatial-Temporal Attention Residual Temporal Convolutional Network (STA-Res-TCN) for skeleton-based dynamic hand gesture recognition, which learns different levels of attention and assigns them to each spatialtemporal feature extracted by the convolution filters at each time step. The proposed attention branch assists the networks to adaptively focus on the informative time frames and features while exclude the irrelevant ones that often bring in unnecessary noise. Moreover, our proposed STARes -TCN is a lightweight model that can be trained and tested in an extremely short time. Experiments on DHG-14/28 Dataset and SHREC'17 Track Dataset show that STARes -TCN outperforms stateof-the-art methods on both the 14 gestures setting and the more complicated 28 gestures setting.
Small public spaces are the key built environment elements that provide venues for various of activities. However, existing measurements or approaches could not efficiently and effectively quantify how small public spaces are being used. In this paper, we utilized a deep convolutional neural network to quantify the usage of small public spaces through recorded videos as a reliable and robust method to bridge the literature gap. To start with, we deployed photographic devices to record videos that cover the minimum enclosing square of a small public space for a certain period of time, then utilized a deep convolutional neural network to detect people in these videos and converted their location from image-based position to real-world projected coordinates. To validate the accuracy and robustness of the method, we experimented our approach in a residential community in Beijing, and our results confirmed that the usage of small public spaces could be measured and quantified effectively and efficiently.
Urban redevelopment is the reconstruction or upgrade of current urban built-up areas; it revitalizes old towns and contributes to sustainable development. This paper proposes a methodological framework that integrates open-source street networks and point-of-interest data and aims to identify and evaluate urban redevelopment at the block level from the perspective of urban form and function. It is found that (1) urban blocks can be categorized into eight groups regarding the spatial form of road junctions that have emerged within them over time, and blocks of each group share common features that can be automatically identified; (2) there are more blocks that have been morphologically redeveloped than functionally redeveloped, and the two types of redevelopments also significantly overlap with one another; and (3) the evaluation of urban redevelopment identification results presents a high accuracy rate that verifies the validity of the proposed framework. Based on the identification results, the impact factors of urban redevelopment are explored on both the inter- and intracity levels. The intercity analysis indicates that Chinese cities with a lower administrative level, lower urbanization rate, and higher density of road junctions tend to be associated with a higher proportion of urban redevelopment. Meanwhile, the intracity analysis attempts to determine which kinds of urban blocks are more likely to undergo urban redevelopment, which are found to be the blocks with lower points of interest density, a smaller distance to city centers, higher transit accessibility, a higher land use mixed index, and larger size.
The effectiveness of runoff control infrastructure depends on infrastructure arrangement and the severity of the problem in the study area. Green infrastructure (GI) has been widely demonstrated as a practical approach to runoff reduction and ecological improvement. However, decision-makers usually consider the cost-efficacy of the GI layout scheme as a primary factor, leading to less consideration of GI’s environmental and ecological functions. Thus, a multifunctional decision-making framework for evaluating the suitability of GI infrastructure was established. First, the study area was described by regional pollution load intensity, slope, available space, and constructible area. Then, to assess the multifunctional performance of GI, a hierarchical evaluation framework comprising three objectives, seven indices, and sixteen sub-indices was established. Weights were assigned to different indices according to stakeholders’ preferences, including government managers, researchers, and residents. The proposed framework can be extended to other cities to detect GI preference.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.