Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Urban Envelope: Representing Audio and Visual Features Extracted from Deep Learning Techniques
https://orcid.org/http://orcid.org/0000-0001-6300-2861
https://orcid.org/http://orcid.org/0000-0001-8210-1566
The study focuses on automated classification and representation of environmental features present in the surroundings. The aim of the study is twofold. Firstly, it compiles a set of Deep Learning (DL) algorithms that are capable of extracting features from street collected images and audio clips. Secondly, it proposes a graphical method to represent such features. The method explains the extracted audio and visual features in a simple illustration that can be automatically created and displayed by analyzing a pair of photographs and audio clips for a particular location. To achieve the objectives, camera and audio recorders are used to collect the street view photographs and sound clips at several locations. These clips are then processed by algorithms to obtain detailed information from such collection. At a given location, auditory and visual features are combined to represent the envelope of sensory information.
Urban Envelope: Representing Audio and Visual Features Extracted from Deep Learning Techniques
https://orcid.org/http://orcid.org/0000-0001-6300-2861
https://orcid.org/http://orcid.org/0000-0001-8210-1566
The study focuses on automated classification and representation of environmental features present in the surroundings. The aim of the study is twofold. Firstly, it compiles a set of Deep Learning (DL) algorithms that are capable of extracting features from street collected images and audio clips. Secondly, it proposes a graphical method to represent such features. The method explains the extracted audio and visual features in a simple illustration that can be automatically created and displayed by analyzing a pair of photographs and audio clips for a particular location. To achieve the objectives, camera and audio recorders are used to collect the street view photographs and sound clips at several locations. These clips are then processed by algorithms to obtain detailed information from such collection. At a given location, auditory and visual features are combined to represent the envelope of sensory information.
Urban Envelope: Representing Audio and Visual Features Extracted from Deep Learning Techniques
https://orcid.org/http://orcid.org/0000-0001-6300-2861
https://orcid.org/http://orcid.org/0000-0001-8210-1566
The study focuses on automated classification and representation of environmental features present in the surroundings. The aim of the study is twofold. Firstly, it compiles a set of Deep Learning (DL) algorithms that are capable of extracting features from street collected images and audio clips. Secondly, it proposes a graphical method to represent such features. The method explains the extracted audio and visual features in a simple illustration that can be automatically created and displayed by analyzing a pair of photographs and audio clips for a particular location. To achieve the objectives, camera and audio recorders are used to collect the street view photographs and sound clips at several locations. These clips are then processed by algorithms to obtain detailed information from such collection. At a given location, auditory and visual features are combined to represent the envelope of sensory information.
Urban Envelope: Representing Audio and Visual Features Extracted from Deep Learning Techniques
Lecture Notes in Civil Engineering
Jana, Arnab (Herausgeber:in) / Banerji, Pradipta (Herausgeber:in) / Verma, Deepank (Autor:in) / Jana, Arnab (Autor:in)
18.03.2021
9 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Representing buildings and urban features in hydrodynamic flood models
| Wiley | 2024
Representing buildings and urban features in hydrodynamic flood models
| Wiley | 2024