6 – 02 Brightness Constancy Assumption V3

Now, optical flow assumes that points in one image frame have the same intensity pixel value as those same points in the next image frame. That is optical flow assumes that the color of a surface will stay the same over time. In practice, this is not a perfect assumption but it’s close most of … Read more

5 – 01 Motion Vector V2

So, how exactly does optical flow work? We know that it tries to track points from one image frame to another based on the intensity levels of points in each image. Let’s see this in a simple example. Say we have two image frames from a video, and for one point on object in image … Read more

4 – Optical Flow

Optical flow is used in many tracking and motion analysis applications. It works by assuming two things about image frames. One, that the pixel intensities of an object do not change between consecutive frames and two, that neighboring pixels have similar motion. It then looks at interesting points, say, corners or particularly bright pixels, and … Read more

3 – Motion

So far we’ve been looking at a lot of static images and processing them to identify objects and interesting features, and the exact same processing techniques can be used on a video stream. This is because a video stream is just made up of a sequence of image frames. One thing we haven’t talked about … Read more

2 – Localization V1 (1)

I’ll be teaching this section in tandem with Sebastian Thrun, who’s been called “The Grandfather of Self-driving cars”. We’ll start off by talking about representing motion and tracking objects in a video. Then, we’ll move onto uncertainty and robotic motion and learn a simple localization technique, the histogram filter, which can account for this uncertainty. … Read more

1 – Object Tracking V4 (1)

So far, you’ve learned a lot about how to extract important information from images and how to use deep learning techniques to recognize patterns in images and identify objects. In this part of the course, we’ll focus less on images and more on pattern recognition over time and space. We’ll talk about different computer vision … Read more