AI6131 Project Description
Overview As part of the 3D Deep Learning course, students are required to complete a project that allows them to explore a topic of their choice related to 3D deep learning. This project is designed to encourage creativity, independent research, and hands-on implementation of 3D deep learning techniques. Students may select any topic within the broad field of 3D deep learning and encouraged to propose novel ideas, apply existing techniques to new problems, or enhance prior work. Assessment will be based on their ability to identify and address challenges, present their findings eƯ ectively, and document their results in a structured manner, reflecting the review process of a top-tier conference submission but with less stringent criteria.
Assessment Components The project consists of three key deliverables:
1. Project Proposal (Due: 19 March)
Length: 1-page document
The proposal should outline: 1) The problem being addressed. 2) The motivation for the project. 3) Key challenges involved. 4) The approach and plan for implementation.
The instructor will review proposals and provide feedback.
2. Project Presentation (16 & 17 April)
Duration: 10-minute presentation per student.
The presentation should cover: 1) The problem and motivation. 2) The methodology used. 3) Preliminary findings and progress. 4) Any challenges encountered and potential solutions.
This session provides an opportunity for direct interaction with the instructor, allowing for feedback and discussion.
3. Final Report (Due: 30 April)
Length: 4-6 pages, structured like a short research paper.
Suggested format:
Title & Abstract: A concise summary of the work
Introduction: Problem statement, motivation, and background
Related Work: Overview of relevant prior research
Methodology: Explanation of the approach and techniques used
Results: Description of experiments, evaluation metrics, and findings
Discussions: Interpretation of results, and possible improvements
Conclusion & Future Work: Summary and potential extensions
Like submitting a paper to a top-tier conference, an optional video demonstration is encouraged.
Additional Notes
Collaboration is not allowed; each student must complete their own project.
Students may use open-source code and publicly available datasets but must clearly acknowledge them in the report. Plagiarism in any form. will not be tolerated.
Students may use Large Language Models (LLMs) to refine the writing in their report. However, they must provide their own insights and findings. Directly generating the report using LLMs is strictly prohibited.
Students are encouraged to seek guidance from the instructor during oƯ ice hours if needed.
This project is an opportunity for students to deepen their understanding of 3D deep learning and gain hands-on experience in conducting research and experimentation. I look forward to seeing your ideas and results!