Homework 1

Submission date: December 14th, 2019

Topics

• Continuous and discrete 2D signals
• Fourier transform
• Sampling
• Burst imaging
• Image deblurring

Downloading

The assignment is available here.

We recommend you use git to clone the repo:

git clone https://github.com/vistalab-technion/cs236860-hw.git

This will allow you to pull updates from us in the event that they are needed.

FAQ

Q: What does it mean that the camera optics can be assumed to be an ideal anti-aliasing filter?

A: It means that the PSF, $h_k$, acts as an ideal anti-aliasing filter according to our sampling grid.

Q: Do we need to compute the integral in question 4?

A: No. You can find the upper bound without computing the integral.

Q: Do we need to compute the integral in questions 5 & 6?

A: Yes. We have seen similar integrals in the lecture.

Q: In question 7, what does it mean that the frames are pre-aligned?

A: We can assume that $o\left(\left(k+0.5\right)T\right)=0$ for each $k$, meaning that the camera is always located in the center of the scene in the middle of its movement during which the frame is taken.

Q: What is considered a valid answer for question 7?

A: Any reasonable algorithm that uses the mathematical foundation we have constructed during the previous questions. An algorithm without an explanation will not get any points.

Q: Can we assume we know $h_k$ or $P_k$ in question 7 and/or in Part II?

A: No. You are only given the blurred images $g_k$.