Search

Crowd flow prediction has been a significant problem in smart cities. Real‐world datasets are the basis of designing data‐driven models, for example, machine learning and deep learning models. However, some well‐known datasets, for example, TaxiBJ, become unavailable when the original sharing links are deleted as the data providers move to new jobs. As an alternative choice, TaxiBJ21 is proposed in this letter for future crowd flow prediction studies. This new open dataset contains the taxi inflow and outflow matrices in Beijing for 3 months, with the same data format as TaxiBJ and can be used in the current literature seamlessly. In this letter, the process of constructing TaxiBJ21 is introduced and three baselines including two simple historical models and a multilayer perceptron neural network are evaluated on this new dataset as benchmarks. Both the data and baselines are publicly available in a Github repository. [ABSTRACT FROM AUTHOR]

The accuracy of biometric recognition in unconstrained scenarios has been a major concern for a large number of researchers. Despite such efforts, no system can recognise in a fully automated manner human beings in totally wild conditions such as in surveillance environments. Several sets of degraded data have been made available to the research community, where the reported performance by state‐of‐the‐art algorithms is already saturated, suggesting that these sets do not reflect faithfully the conditions in such hard settings. To this end, the authors introduce the QUIS‐CAMPI data feed, comprising samples automatically acquired by an outdoor visual surveillance system, with subjects on‐the‐move and at‐a‐distance (up to 50 m). Also, they supply a high‐quality set of enrolment data. When compared to similar data sources, the major novelties of QUIS‐CAMPI are: (i) biometric samples are acquired in a fully automatic way; (ii) it is an open dataset, i.e. the number of probe images and enroled subjects grow on a daily basis; and (iii) it contains multi‐biometric traits. The ensemble properties of QUIS‐CAMPI ensure that the data span a representative set of covariate factors of real‐world scenarios, making it a valuable tool for developing and benchmarking biometric recognition algorithms capable of working in unconstrained scenarios. [ABSTRACT FROM AUTHOR]

Purpose: We describe a public dataset with MR and CT images of patients performed in the same position with both multiobserver and expert consensus delineations of relevant organs in the male pelvic region. The purpose was to provide means for training and validation of segmentation algorithms and methods to convert MR to CT like data, i.e., so called synthetic CT (sCT). Acquisition and validation methods: T1‐ and T2‐weighted MR images as well as CT data were collected for 19 patients at three different departments. Five experts delineated nine organs for each patient based on the T2‐weighted MR images. An automatic method was used to fuse the delineations. Starting from each fused delineation, a consensus delineation was agreed upon by the five experts for each organ and patient. Segmentation overlap between user delineations with respect to the consensus delineations was measured to describe the spread of the collected data. Finally, an open‐source software was used to create deformation vector fields describing the relation between MR and CT images to further increase the usability of the dataset. Data format and usage notes: The dataset has been made publically available to be used for academic purposes, and can be accessed from https://zenodo.org/record/583096. Potential applications: The dataset provides a useful source for training and validation of segmentation algorithms as well as methods to convert MR to CT‐like data (sCT). To give some examples: The T2‐weighted MR images with their consensus delineations can directly be used as a template in an existing atlas‐based segmentation engine; the expert delineations are useful to validate the performance of a segmentation algorithm as they provide a way to measure variability among users which can be compared with the result of an automatic segmentation; and the pairwise deformably registered MR and CT images can be a source for an atlas‐based sCT algorithm or for validation of sCT algorithm. [ABSTRACT FROM AUTHOR]