Hello, if you have any need, please feel free to consult us, this is my wechat: wx91due
CIS125 Project (30% of Total Marks)
The project consists of 2 parts, both of which have to be attempted. These are meant to be assignment-like and supplement the labs and quizzes conducted in class.
Part A (50 marks)
Reference: Lab in Lecture 8 (but no need to carry out the steps in this Lab)
A.1 (40 marks)
- Select 4 image processing filters from this list: Gotham, Sepia, pencil sketch, sharpening, detail enhancing, inverting, painting-stylizing.
- Implement the image filters using OpenCV library functions showing original and transformed images (after applying one of the above filters selected by you).
- Explain the working principle of each filter in a paragraph, highlighting the underlying concepts and algorithms in the report (NO need to explain in the program).
A.2 – advanced (10 marks)
- Research and understand the general concept of applying deep learning techniques for style transfer, specifically transferring artistic styles like impressionist or Van Gogh to images.
- Explain the process of style transfer techniques to images briefly with illustrations and examples.
Marking Rubric
A.1 (40 marks)
- Successfully implemented and demonstrated the image filters (16 marks)
- Explained the working principle of each filter in easy-to-understand language (24 marks)
A.2 (10 marks)
- Explained the process of style transfer techniques with illustrations and images in easy-to-understand language and some level of technical details (No program needed on this part).
Part B (50 marks)
Reference: Lab in Lecture 11
Record with your cell phone or computer or find 2 sound tracks in .wav format between 30 to 60 seconds. The sound tracks can be in either ‘Mono’ or ‘Stereo’ (though ‘Mono’ allows you to deal with only 1 track only). Track1 can be recordings of speech, news broadcast or a person singing. Track2 should be background music or musical instrument accompaniment.
Duplicate the two tracks (so you have 4 tracks in your project). Change names of the original tracks to indicate they are the originals. Turn these original tracks to ‘mute’.
For each of the two duplicated tracks, go through the 4 step exercise shown in class in this order:
1. Noise Reduction
2. Equalization
3. Compressor
4. Normalize
Add other effects that you may see fit or may add interesting results to the tracks. Save the entire project in an Audacity project file format (.aup3 ). Submit this project file.
Marking Rubric:
- Correct implementation of the 4 steps in the correct order to enhance the two tracks (15 marks)
- Proper description of steps taken, reasons for the choice of parameters in each step and the ‘perceived’ improvement to the original tracks (15 marks)
- Additional effects (not including those 4 listed above) that added interesting qualities to the tracks (10 marks)
- Presentation of pleasant and creative materials in the sound tracks when played together (10 marks)
Submission:
A. Report (PDF)
- For A.1, complete the code implementation for the 4 image processing filters chosen using OpenCV and Python in a shared ipynb file (with public access). Paste the link in the report.
- Provide the working principle explanations for each filter in A.1 and the process of style transfer in A.2 in the report
- For Part B, list and describe all the steps and parameters you have taken to modify the tracks and the respective reasons for the choice of parameters, and a paragraph explaining what are your ‘perceived’ improvement to the tracks as compared to the original tracks after applying the enhancements
- Submit this report to BOTH the Turnitin submission and the Project submission window
- The submission to the Turnitin window is for checking of use of Generative AI and other public content. If there are problems with embedded images and links during this submission, take out the images and links and just submit the text report to the Turnitin window. The full report should still be submitted to the other Project window for full grading.
B. Audacity Project File (.aup3)
- The Audacity project file that contains the 4 tracks: 2 original and 2 modified as instructed above to be submitted to the Project submission window
- Note: all the tracks should be in the SAME project file, not separate project files• If the total file size exceeds 100MB, shorten the duration of the tracks to reduce the
- Audacity project file size
NOTES:
- Do not zip your files when submitting. Just submit the Report and Audacity files with clearly indicated names
- Generative AI tools are allowed in a limited sense. It is crucial for students to demonstrate their understanding of the material and concepts involved. If a student's work exhibits a Turnitin similarity score exceeding 40%, 40 marks (out of a total 100) will be deducted.
- In order to maintain fairness and uphold academic integrity, it is explicitly stated that if two students' project submissions demonstrate a high level of similarity, 40 marks will be deducted from both students' overall marks. Each student is expected to independently develop and present their work, reflecting their own understanding and efforts. Collaboration and discussion among students are encouraged, but the final project submissions should showcase individuality and unique contributions.
- The project maintains a zero-tolerance policy towards the use of offensive or abusive image, video, or audio materials. Any inclusion or dissemination of content that is offensive, discriminatory, harassing, or harmful to individuals or groups based on factors such as race, ethnicity, gender, religion, nationality, or any other protected characteristic will be met with severe consequences including but not limited to receiving 0 marks for this entire project.