The VideoSpark platform, as part of the Motork Product Suite, is a groundbreaking software leveraging artificial intelligence for the production of automotive videos and commercials. Its primary aim is to simplify the content creation process, minimize reliance on extensive resources or expertise, and reduce marketing costs, all while enhancing conversion rates through the creation of captivating and informative videos. The process involves collecting vehicle text data and images, sorting these images and organizing vehicle information for optimal display, and then automatically generating advertising videos using templates and pre-defined soundtracks. The utilization of AI models, including Convolutional Neural Networks (CNNs), Support Vector Machine (SVM) for image classification, angle detection, and the YOLO algorithm for object detection and segmentation, ensures that the most marketing-relevant features of the vehicle are showcased. Decision trees further aid in classifying textual data into coherent categories, enriching the videos with relevant information. The final step involves integrating the classified and processed image and text data to create engaging videos, which are then distributed across various digital platforms. The project's infrastructure and user interface are designed for scalability and reliability, employing technologies such as Python, FastAPI, Docker, SQLAlchemy, VueJs, and Tailwind CSS. This thesis outlines an innovative approach to automotive marketing through AI, aiming to redefine traditional content creation methods and establish new benchmarks for efficiency, accessibility, and impact. Its applications span various industries such as real estate, tourism, e-commerce, education, and healthcare, showcasing its wide-reaching potential and versatility.
The VideoSpark platform, as part of the Motork Product Suite, is a groundbreaking software leveraging artificial intelligence for the production of automotive videos and commercials. Its primary aim is to simplify the content creation process, minimize reliance on extensive resources or expertise, and reduce marketing costs, all while enhancing conversion rates through the creation of captivating and informative videos. The process involves collecting vehicle text data and images, sorting these images and organizing vehicle information for optimal display, and then automatically generating advertising videos using templates and pre-defined soundtracks. The utilization of AI models, including Convolutional Neural Networks (CNNs), Support Vector Machine (SVM) for image classification, angle detection, and the YOLO algorithm for object detection and segmentation, ensures that the most marketing-relevant features of the vehicle are showcased. Decision trees further aid in classifying textual data into coherent categories, enriching the videos with relevant information. The final step involves integrating the classified and processed image and text data to create engaging videos, which are then distributed across various digital platforms. The project's infrastructure and user interface are designed for scalability and reliability, employing technologies such as Python, FastAPI, Docker, SQLAlchemy, VueJs, and Tailwind CSS. This thesis outlines an innovative approach to automotive marketing through AI, aiming to redefine traditional content creation methods and establish new benchmarks for efficiency, accessibility, and impact. Its applications span various industries such as real estate, tourism, e-commerce, education, and healthcare, showcasing its wide-reaching potential and versatility.
Driving Change: AI-Powered Content Creation for Next-Gen Automotive Marketing
TALEBI, MOHAMMADMOSTAFA
2023/2024
Abstract
The VideoSpark platform, as part of the Motork Product Suite, is a groundbreaking software leveraging artificial intelligence for the production of automotive videos and commercials. Its primary aim is to simplify the content creation process, minimize reliance on extensive resources or expertise, and reduce marketing costs, all while enhancing conversion rates through the creation of captivating and informative videos. The process involves collecting vehicle text data and images, sorting these images and organizing vehicle information for optimal display, and then automatically generating advertising videos using templates and pre-defined soundtracks. The utilization of AI models, including Convolutional Neural Networks (CNNs), Support Vector Machine (SVM) for image classification, angle detection, and the YOLO algorithm for object detection and segmentation, ensures that the most marketing-relevant features of the vehicle are showcased. Decision trees further aid in classifying textual data into coherent categories, enriching the videos with relevant information. The final step involves integrating the classified and processed image and text data to create engaging videos, which are then distributed across various digital platforms. The project's infrastructure and user interface are designed for scalability and reliability, employing technologies such as Python, FastAPI, Docker, SQLAlchemy, VueJs, and Tailwind CSS. This thesis outlines an innovative approach to automotive marketing through AI, aiming to redefine traditional content creation methods and establish new benchmarks for efficiency, accessibility, and impact. Its applications span various industries such as real estate, tourism, e-commerce, education, and healthcare, showcasing its wide-reaching potential and versatility.File | Dimensione | Formato | |
---|---|---|---|
Mohammadmostafa_Talebi.pdf
embargo fino al 07/03/2025
Dimensione
3.07 MB
Formato
Adobe PDF
|
3.07 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/62377