Python代写 | Computer Vision Coursework Assignment


1 Assignment Overview
This assignment will involve you designing, building, testing and critiquing a system for per-
forming face alignment, aka. locating facial landmarks in images. There are also some small
extension tasks detailed below.
It is worth 80% of the grade for this module. It’s designed to ensure you can demonstrate
achieving the learning outcomes for this module, which are:
• Write and document a computer program to extract useful information from image data.
• Propose designs for simple computer vision systems.
• Determine the applicability of a variety of computer vision techniques to practical prob-
• Describe and recognise the e ects of a variety of image processing operations.
1.1 Extension Tasks
1. Provide a simple solution for segmenting the pixels of the face and/or skin (10%).
2. Implement a simple graphical e ect that uses the information you have extracted from the
image (10%).

2 What to hand in?
1. A report that comprises a maximum of 8 pages and 1500 words, including captions but
excluding references. I’m expecting several pictures, diagrams, owcharts and charts to be
• A summary and justi cation for all the steps in your face alignment system, including
preprocessing, choice of image features and prediction model. Explaining diagram-
matically is very welcome.
• Results of your experiments: This should include some discussion of qualitative (ex-
ample based) and quantitative (number based) comparisons between di erent ap-
proaches that you have experimented with.
• A short summary of your face segmentation system and graphics e ects methodology.
• Qualitative examples of your methods running on the small set of provided example
images, found in the compressed numpy le (examples.npz) here. Include pictures of
the result of face alignment, face segmentation and graphical e ects systems.

• Examples of failure cases in the face alignment system and a critical analysis of these,
identifying potential causes and solutions.
2. A CSV le that contains the face landmark positions on the test set of images, found in the
compressed numpy le (\test images”.npz) here. Use the provided \save as csv” function
in the colab worksheet to process an array of shape (number test image, number points,
2) to a csv le.
3. Either links to Colab worksheets, .ipynb les, .py les containing annotated code for any
data preprocessing, model training and testing.
3 How will this be graded?
The breakdown of marks for this assignment are given below:
25 Marks Accuracy and robustness of face alignment
These marks are allocated based on the performance of the face alignment method. This
will be evaluated on the held out test set, which includes some dicult cases. The test
images, without annotations are provided in the compressed numpy le (test images.npz)
here and the error on the predicted points will be calculated after submission. Marks will
be awarded for average accuracy and robustness (% of images with error below a certain
threshold) as well as qualitative evaluation on the example images.

30 Marks Outline of methods employed
Justifying and explaining design decisions for the landmark nding. This does not have to
be in depth, and I do not expect you to regurgitate the contents of the lecture notes/papers.
You should state clearly:
• what methods you have used, with what training parameters and why.
• what image features you have used, brie y describe how they were calculated, and
why you chose them.
• any image pre-processing steps you have used, and why.
For top marks, you should clearly demonstrate a creative and methodical approach for
designing your system, drawing ideas from di erent sources. Explaining using diagrams
and/or owcharts is very welcome.
20 Marks Analysing results and failure cases
Critically evaluate the results produced by your system on test/validation data. You should
include quantitative (number based) and qualitative (example based) comparisons between
di erent approaches that you have tried. Quantitative measures including measuring the
cumulative error distribution (see lecture slides) or using boxplots or other plots to compare
methods. Please note that I am interested in your nal prediction results, rather than how
the cost function changes during training. A detailed qualitative analysis would investigate
and identify systematic failure cases, providing visual examples, and propose potential
10 Marks Face and/or skin segmentation
Very brie y outline/provide a diagram/ owchart explaining how this was performed. Pro-
vide some qualitative results and illustrate some failure cases. I am only expecting very