Gait Recognition by Multidimensional Representation for Accumulated Silhouette

Author(s):  
Seungdo Jeong ◽  
Jungwon Cho
2008 ◽  
Author(s):  
Ingrid R. Johnsen ◽  
William A. Cunningham ◽  
John B. Nezlek

2004 ◽  
Author(s):  
Zongyi Liu ◽  
Laura Malave ◽  
Adebola Osuntogun ◽  
Preksha Sudhakar ◽  
Sudeep Sarkar
Keyword(s):  

Buildings ◽  
2021 ◽  
Vol 11 (2) ◽  
pp. 78
Author(s):  
Daria Uspenskaia ◽  
Karl Specht ◽  
Hendrik Kondziella ◽  
Thomas Bruckner

Without decarbonizing cities energy and climate objectives cannot be achieved as cities account for approximately two thirds of energy consumption and emissions. This goal of decarbonizing cities has to be facilitated by promoting net-zero/positive energy buildings and districts and replicating them, driving cities towards sustainability goals. Many projects in smart cities demonstrate novel and groundbreaking low-carbon solutions in demonstration and lighthouse projects. However, as the historical, geographic, political, social and economic context of urban areas vary greatly, it is not always easy to repeat the solution in another city or even district. It is therefore important to look for the opportunities to scale up or repeat successful pilots. The purpose of this paper is to explore common trends in technologies and replication strategies for positive energy buildings or districts in smart city projects, based on the practical experience from a case study in Leipzig—one of the lighthouse cities in the project SPARCS. One of the key findings the paper has proven is the necessity of a profound replication modelling to deepen the understanding of upscaling processes. Three models analyzed in this article are able to provide a multidimensional representation of the solution to be replicated.


2020 ◽  
pp. 1-12
Author(s):  
Linuo Wang

The current technology related to athlete gait recognition has shortcomings such as complicated equipment and high cost, and there are also certain problems in recognition accuracy and recognition efficiency. In order to improve the efficiency of athletes’ gait recognition, this paper studies the different recognition technologies of athletes based on machine learning and spectral feature technology and applies computer vision technology to sports. Moreover, according to the calf angular velocity signal, the occurrence of leg movement is detected in real time, and the gait cycle is accurately divided to reduce the influence of the signal unrelated to the behavior on the recognition process. In addition, this study proposes a gait behavior recognition method based on event-driven strategies. This method uses a gyroscope as the main sensor and uses a wearable sensor node to collect the angular velocity signals of the legs and waist. In addition, this study analyzes the performance of the algorithm proposed by this paper through experimental research. The comparison results show that the method proposed by this paper has improved the number of recognition action types and accuracy and has certain advantages from the perspective of computation and scalability.


Sign in / Sign up

Export Citation Format

Share Document