A two-stream network with joint spatial-temporal distance for video-based person re-identification

2020 ◽  
Vol 39 (3) ◽  
pp. 3769-3781
Author(s):  
Zhisong Han ◽  
Yaling Liang ◽  
Zengqun Chen ◽  
Zhiheng Zhou
Keyword(s):  
Spatial Information ◽  
Temporal Information ◽  
Temporal Distance ◽  
Spatial Distance ◽  
Weighted Sum ◽  
Stream Network ◽  
Spatial Features ◽  
Public Datasets ◽  
Types Of Information ◽  
Temporal Information Extraction

Video-based person re-identification aims to match videos of pedestrians captured by non-overlapping cameras. Video provides spatial information and temporal information. However, most existing methods do not combine these two types of information well and ignore that they are of different importance in most cases. To address the above issues, we propose a two-stream network with a joint distance metric for measuring the similarity of two videos. The proposed two-stream network has several appealing properties. First, the spatial stream focuses on multiple parts of a person and outputs robust local spatial features. Second, a lightweight and effective temporal information extraction block is introduced in video-based person re-identification. In the inference stage, the distance of two videos is measured by the weighted sum of spatial distance and temporal distance. We conduct extensive experiments on four public datasets, i.e., MARS, PRID2011, iLIDS-VID and DukeMTMC-VideoReID to show that our proposed approach outperforms existing methods in video-based person re-ID.

scholarly journals Visible-Light Camera Sensor-Based Presentation Attack Detection for Face Recognition by Combining Spatial and Temporal Information

Sensors ◽  
2019 ◽  
Vol 19 (2) ◽  
pp. 410 ◽  
Author(s):  
Dat Nguyen ◽  
Tuyen Pham ◽  
Min Lee ◽  
Kang Park
Keyword(s):  
Neural Network ◽  
Visible Light ◽  
Spatial Information ◽  
Recognition System ◽  
Image Features ◽  
Attack Detection ◽  
Temporal Information ◽  
Public Datasets ◽  
Presentation Attack Detection ◽  
Camera Sensor

Face-based biometric recognition systems that can recognize human faces are widely employed in places such as airports, immigration offices, and companies, and applications such as mobile phones. However, the security of this recognition method can be compromised by attackers (unauthorized persons), who might bypass the recognition system using artificial facial images. In addition, most previous studies on face presentation attack detection have only utilized spatial information. To address this problem, we propose a visible-light camera sensor-based presentation attack detection that is based on both spatial and temporal information, using the deep features extracted by a stacked convolutional neural network (CNN)-recurrent neural network (RNN) along with handcrafted features. Through experiments using two public datasets, we demonstrate that the temporal information is sufficient for detecting attacks using face images. In addition, it is established that the handcrafted image features efficiently enhance the detection performance of deep features, and the proposed method outperforms previous methods.

scholarly journals A Robust Lane Detection Model Using Vertical Spatial Features and Contextual Driving Information

Sensors ◽  
2021 ◽  
Vol 21 (3) ◽  
pp. 708
Author(s):  
Wenbo Liu ◽  
Fei Yan ◽  
Jiyong Zhang ◽  
Tao Deng
Keyword(s):  
Information Exchange ◽  
Information Transfer ◽  
Spatial Information ◽  
Contextual Information ◽  
Great Influence ◽  
Light Condition ◽  
Lane Detection ◽  
Detection Model ◽  
Spatial Features ◽  
Proposed Model

The quality of detected lane lines has a great influence on the driving decisions of unmanned vehicles. However, during the process of unmanned vehicle driving, the changes in the driving scene cause much trouble for lane detection algorithms. The unclear and occluded lane lines cannot be clearly detected by most existing lane detection models in many complex driving scenes, such as crowded scene, poor light condition, etc. In view of this, we propose a robust lane detection model using vertical spatial features and contextual driving information in complex driving scenes. The more effective use of contextual information and vertical spatial features enables the proposed model more robust detect unclear and occluded lane lines by two designed blocks: feature merging block and information exchange block. The feature merging block can provide increased contextual information to pass to the subsequent network, which enables the network to learn more feature details to help detect unclear lane lines. The information exchange block is a novel block that combines the advantages of spatial convolution and dilated convolution to enhance the process of information transfer between pixels. The addition of spatial information allows the network to better detect occluded lane lines. Experimental results show that our proposed model can detect lane lines more robustly and precisely than state-of-the-art models in a variety of complex driving scenarios.

scholarly journals The Role of Spatio-Temporal Information to Govern the COVID-19 Pandemic: A European Perspective

2021 ◽  
Vol 10 (3) ◽  
pp. 166
Author(s):  
Hartmut Müller ◽  
Marije Louwsma
Keyword(s):  
European Union ◽  
Spatial Distribution ◽  
Spatial Information ◽  
National Level ◽  
Temporal Information ◽  
The European Union ◽  
Member States ◽  
Heavy Burden ◽  
Spatio Temporal

The Covid-19 pandemic put a heavy burden on member states in the European Union. To govern the pandemic, having access to reliable geo-information is key for monitoring the spatial distribution of the outbreak over time. This study aims to analyze the role of spatio-temporal information in governing the pandemic in the European Union and its member states. The European Nomenclature of Territorial Units for Statistics (NUTS) system and selected national dashboards from member states were assessed to analyze which spatio-temporal information was used, how the information was visualized and whether this changed over the course of the pandemic. Initially, member states focused on their own jurisdiction by creating national dashboards to monitor the pandemic. Information between member states was not aligned. Producing reliable data and timeliness reporting was problematic, just like selecting indictors to monitor the spatial distribution and intensity of the outbreak. Over the course of the pandemic, with more knowledge about the virus and its characteristics, interventions of member states to govern the outbreak were better aligned at the European level. However, further integration and alignment of public health data, statistical data and spatio-temporal data could provide even better information for governments and actors involved in managing the outbreak, both at national and supra-national level. The Infrastructure for Spatial Information in Europe (INSPIRE) initiative and the NUTS system provide a framework to guide future integration and extension of existing systems.

scholarly journals Psychological distance modulates goal-based versus movement-based imitation

2019 ◽  
Author(s):  
Oliver Genschow ◽  
Jochim Hansen ◽  
Michaela Wänke ◽  
Yaacov Trope
Keyword(s):  
Pilot Study ◽  
Psychological Distance ◽  
Past Research ◽  
Construal Level Theory ◽  
Temporal Distance ◽  
Spatial Distance ◽  
Construal Level ◽  
Direct Mapping

In past research on imitation, some findings suggest that imitation is goal based, whereas other findings suggest that imitation can also be based on a direct mapping of a model’s movements without necessarily adopting the model’s goal. We argue that the two forms of imitation are flexibly deployed in accordance with the psychological distance from the model. We specifically hypothesize that individuals are relatively more likely to imitate the model’s goals when s/he is distant but relatively more likely to imitate the model’s specific movements when s/he is proximal. This hypothesis was tested in four experiments using different imitation paradigms and different distance manipulations. Experiment 1 served as a pilot study and demonstrated that temporal distance (vs. proximity) increased imitation of a goal relative to the imitation of a movement. Experiments 2 and 3 measured goal-based and movement-based imitation independently of each other and found that spatial distance (vs. proximity) decreased the rate of goal errors (indicating more goal imitation) compared to movement errors. Experiment 4 demonstrated that psychological distance operates most likely at the input—that is, perceptual—level. The findings are discussed in relation to construal level theory and extant theories of imitation.

scholarly journals A time-stamp mechanism may provide temporal information necessary for egocentric to allocentric spatial transformations

eLife ◽  
2018 ◽  
Vol 7 ◽  
Author(s):  
Avner Wallach ◽  
Erik Harvey-Girard ◽  
James Jaeyoon Jun ◽  
André Longtin ◽  
Len Maler
Keyword(s):  
Spatial Learning ◽  
Spatial Organization ◽  
Spatial Information ◽  
Spatial Relations ◽  
Temporal Information ◽  
Accurate Estimation ◽  
Neural Recordings ◽  
Brain Circuitry ◽  
Learning Circuits ◽  
Thalamic Region

Learning the spatial organization of the environment is essential for most animals’ survival. This requires the animal to derive allocentric spatial information from egocentric sensory and motor experience. The neural mechanisms underlying this transformation are mostly unknown. We addressed this problem in electric fish, which can precisely navigate in complete darkness and whose brain circuitry is relatively simple. We conducted the first neural recordings in the preglomerular complex, the thalamic region exclusively connecting the optic tectum with the spatial learning circuits in the dorsolateral pallium. While tectal topographic information was mostly eliminated in preglomerular neurons, the time-intervals between object encounters were precisely encoded. We show that this reliable temporal information, combined with a speed signal, can permit accurate estimation of the distance between encounters, a necessary component of path-integration that enables computing allocentric spatial relations. Our results suggest that similar mechanisms are involved in sequential spatial learning in all vertebrates.

scholarly journals Fatigue Detection with Spatial-Temporal Fusion Method on Covariance Manifolds of Electroencephalography

Entropy ◽  
2021 ◽  
Vol 23 (10) ◽  
pp. 1298
Author(s):  
Nan Zhao ◽  
Dawei Lu ◽  
Kechen Hou ◽  
Meifei Chen ◽  
Xiangyu Wei ◽  
...  
Keyword(s):  
Traffic Accidents ◽  
Covariance Matrices ◽  
Temporal Information ◽  
Pressure Work ◽  
Spatial Features ◽  
Eeg Data ◽  
High Level ◽  
Fatigue Driving ◽  
Fatigue Detection ◽  
Better Than

With the increasing pressure of current life, fatigue caused by high-pressure work has deeply affected people and even threatened their lives. In particular, fatigue driving has become a leading cause of traffic accidents and deaths. This paper investigates electroencephalography (EEG)-based fatigue detection for driving by mining the latent information through the spatial-temporal changes in the relations between EEG channels. First, EEG data are partitioned into several segments to calculate the covariance matrices of each segment, and then we feed these matrices into a recurrent neural network to obtain high-level temporal information. Second, the covariance matrices of whole signals are leveraged to extract two kinds of spatial features, which will be fused with temporal characteristics to obtain comprehensive spatial-temporal information. Experiments on an open benchmark showed that our method achieved an excellent classification accuracy of 93.834% and performed better than several novel methods. These experimental results indicate that our method enables better reliability and feasibility in the detection of fatigued driving.

Time-topology analysis

2021 ◽  
Vol 14 (13) ◽  
pp. 3322-3334
Author(s):  
Yunkai Lou ◽  
Chaokun Wang ◽  
Tiankai Gu ◽  
Hao Feng ◽  
Jun Chen ◽  
...  
Keyword(s):  
Temporal Information ◽  
Time Dimension ◽  
Topology Analysis ◽  
Topological Information ◽  
Analysis Methods ◽  
Short Period ◽  
Pruning Strategy ◽  
Temporal Graph ◽  
Types Of Information ◽  
Temporal Graphs

Many real-world networks have been evolving, and are finely modeled as temporal graphs from the viewpoint of the graph theory. A temporal graph is informative, and always contains two types of information, i.e., the temporal information and topological information, where the temporal information reflects the time when the relationships are established, and the topological information focuses on the structure of the graph. In this paper, we perform time-topology analysis on temporal graphs to extract useful information. Firstly, a new metric named T-cohesiveness is proposed to evaluate the cohesiveness of a temporal subgraph. It defines the cohesiveness of a temporal subgraph from the time and topology dimensions jointly. Specifically, given a temporal graph G s = ( Vs , ε Es ), cohesiveness in the time dimension reflects whether the connections in G s happen in a short period of time, while cohesiveness in the topology dimension indicates whether the vertices in V s are densely connected and have few connections with vertices out of G s . Then, T-cohesiveness is utilized to perform time-topology analysis on temporal graphs, and two time-topology analysis methods are proposed. In detail, T-cohesiveness evolution tracking traces the evolution of the T-cohesiveness of a subgraph, and combo searching finds out all the subgraphs that contain the query vertex and have T-cohesiveness larger than a given threshold. Moreover, a pruning strategy is proposed to improve the efficiency of combo searching. Experimental results confirm the efficiency of the proposed time-topology analysis methods and the pruning strategy.

Skeleton Network Extraction and Analysis on Bicycle Sharing Networks

2020 ◽  
Vol 16 (3) ◽  
pp. 146-167
Author(s):  
Kanokwan Malang ◽  
Shuliang Wang ◽  
Yuanyuan Lv ◽  
Aniwat Phaphuangwittayakul
Keyword(s):  
Spatial Distribution ◽  
Extraction Method ◽  
Spatial Information ◽  
Spatial Clustering ◽  
Transportation Networks ◽  
Network Size ◽  
Network Characteristics ◽  
Spatial Features

Skeleton network extraction has been adopted unevenly in transportation networks whose nodes are always represented as spatial units. In this article, the TPks skeleton network extraction method is proposed and applied to bicycle sharing networks. The method aims to reduce the network size while preserving key topologies and spatial features. The authors quantified the importance of nodes by an improved topology potential algorithm. The spatial clustering allows to detect high traffic concentrations and allocate the nodes of each cluster according to their spatial distribution. Then, the skeleton network is constructed by aggregating the most important indicated skeleton nodes. The authors examine the skeleton network characteristics and different spatial information using the original networks as a benchmark. The results show that the skeleton networks can preserve the topological and spatial information similar to the original networks while reducing their size and complexity.

Sign in / Sign up

Export Citation Format

Share Document