DCU Expo 2023 Final Year Projects

38 36. Creating aNovel Evaluation Technique for Text-To-Video GenerativeModels This project represents a convergence of machine learning, video and image processing, and natural language processing (NLP) in pursuit of a novel and innovative quality evaluation metric for video output from text-to-videomodels. The project combines a custom-built image naturalness and human interpretability classifier with advanced text similarity techniques to produce more precise and dependable performance scores. This initiative employs the latest state-of-the-art image generation and processing technology to enhance user experience and elevate the quality of generated videos. Class Data Science Project Area Artificial Intelligence, Cloud Computing, Computer Vision, Data Analytics, DataMining, Image/Video Processing, Natural Language Processing, Web Application Project Technology Docker, Python, Machine Learning, Git, PaperSpace, SSH into private GPU Student Name(s) Philip Lynch  |  Iya Chivileva Email philip.lynch35@mail.dcu.ie  |  iya.chivileva2@mail.dcu.ie Supervisor Prof TomasWard 37. Modification of a Test Bed for 3DPrinting of FlexibleMaterials This project investigates the repurposing of a 3-axis millingmachine into a functioning 3D printer for flexible materials. Class Mechanical andManufacturing Engineering (Year 4) Project Area 3-DModelling, AdditiveManufacturing Project Technology PLC Programming, Python, Solidworks, G-Code Student Name(s) Ethan Kelly Email ethan.kelly94@mail.dcu.ie Supervisor Dr Nigel Kent 38. SUM-UP This project investigates amethod for automatically generating podcast audio summaries. This project is a web application that concentrates on extractive summaries, which are created by selecting a collection of pertinent audio parts froma podcast without changing or altering the original material. Before listening to the full episode, such summaries would educate the viewer about the podcast’s subjects. The project should point out that the goal is not to offer the viewer a shortened version of the entire show, but rather to give an overview in the style of a film teaser. Class Computer Applications Project Area Multimedia, Natural Language Processing, Speech Recognition, Web Application, Human-computer Interaction Project Technology CSS, HTML5, JavaScript, Python, Machine Learning, Flask, Firebase Student Name(s) Muhammad Zubair Asif  |  Alif Hossain Email muhammad.asif2@mail.dcu.ie   |  alif.hossain5@mail.dcu.ie Supervisor Prof Gareth Jones