Picture this. A senior math major is helping a freshman with calculus. After a few frustrating minutes of “what?”, “OK…”, “hmm” from the freshman the senior gradually becomes more and more animated. He raises his voice to emphasize some important words (“given any epsilon, I can always specify delta…”, “as long as you pick an x, I can guarantee you…”). He is furiously pointing at an especially important spot on the paper (if this is a blackboard, you can hear chalk pounding). He forcefully underlines and circles some important letters and formulas, again and again (if this is a blackboard, the chalk might break at this point). All this time he stares at the freshman with an intense expression on his face.

I always feel like this freshman when I watch Sal Khan’s videos.

Since I heard about the Khan Academy for the first time I’ve been trying to understand the reason for its popularity and why some people think it’s a breakthrough in education. I think I’ve finally figured it out. It’s been right in front of me all this time.

What Sal Khan does is tutoring!

Then everything falls into place:
1. It’s popular as good tutoring would be.
2. It’s a revelation to those who have never tried it.

After a lecture in a room for 300, a ½ hour with a decent tutor will feel much more personal. If the tutor is any good, the student usually becomes a fan, and sometimes a user, of this service — and complains to the tutor about the bad professor. If, on the other hand, the tutor is bad, the student never returns — and complains to the professor about the useless service.

Just about every college offers tutoring services to its students, at least in the US. I always recommend them to my students, but virtually none ever tries it. I suspect this is not uncommon. Good students don’t go there because they feel they don’t need it and bad students because it feels like extra work.

So, once you recognize it for what it is, there is nothing wrong with tutoring style used by Khan. However, building “a new paradigm in education” on this foundation has serious pitfalls. I’ll save that for another time…

Question from a user:

I was looking at your software trial and I am not sure whether or not your software actually calculates the amount of noise in a photo.

The short answer is no. In fact, I’ve never been asked this question before. Typically, users want to extract information from the image. Then Pixcavator gives them the objects largest in size or highest in contrast or both (see Objects in gray scale images).

The answer certainly depends on your definition of the word “noise”. Noise can be measured in a number of ways. Is it the total variation of the color? Specifically, the standard deviation of the gray scale function. If this makes sense to you, ImageJ might help.

This approach however ignores the fact that one has to analyze the image before something is declared noise.

From my point of view, a better approach to noise is to decide first what’s not noise, then look at what’s left.

For example, one can measure noise as the number of objects ignored by analysis. Pixcavator might be able to help here. Two steps: 

Step 1: Find settings that produce meaningful results and capture the important objects in the image.
Step 2: Compare that to the number of objects under very low setting of the sliders.

More here…

Fascinating!

For the geometry behind stereo vision, read this.

I plan to attend IS&T/SPIE Electronic Imaging conference in San Francisco in January 2011. I will be giving a talk at session “Image Processing: Algorithms and Systems IX”. The title is “A graph, non-tree representation of the topology of a gray scale image,” to be presented 25 Jan 2011. Some of the content is here. There will be also a demonstration session that evening. If I am be able to participate, I’ll present the latest version of Pixcavator.