Slashdot Mirror
$6 System-On-A-Chip Mimics Human Vision
Brian McLaughlin writes "This article in TechWeb describes a Visual Perception Processor (costing $6) that can automatically detect objects and track their movement in real time, according to Buereau d'Etudes Vision (BEV). They claim that a full-blown vision processing system/application could be built for less than $50 that rivals current state-of-the-art $10,000 systems. Sounds pretty cool.
" Heck, with my vision, I could tear my eyeballs out and simply use these, at a fraction of the cost of new glasses.
$50 for a reactive vision processing system? Couple that with a cheap (or free) reliable operating system, and cheap networking that has lots of addresses (IPv6), and you could put watching devices on every street corner, heck in every house (you know, for the safety of the children!).
Goodbye privacy.
George
Driving down highway
Blue screen of death suddenly
I crash into tree
If this chip is really as capable as it's made out to be, it will mean a great deal to people who are primarily interested in autonomous mobile robots, as opposed to computer vision.
I could imagine hooking something like this up to a pioneer and solving a bunch of problems.
Sort of makes me wish I were still a student, with the time and resources to play with robots...
The brain-eye system uses layers of parallel-processing neurons that pass the signal through a series of preprocessing steps, resulting in real-time tracking of multiple moving objects within a visual scene.
Get it now? So please guys, don't get any *SMART* ideas and jab your eyes out so you can buy a nice new $6 dollar replacement. (I know no one was actually serious about that, but you guys did still miss the point. Its not the camera thats important, and it would do no good to put xray or a zoom lense on there because you're not looking for the output of what its seeing. Its a TRACKING device. So its not used for monitoring either... geez.) \rant
I am a leaf on the wind. Watch how I soar.
The GVPP's major performance strength over current $10,000 vision systems is its automatic adaptation to varying lighting conditions. Today's vision systems dictate uniform, shadowless illumination, and even next-generation prototype systems, designed to work under "normal" lighting conditions, can be used only from dawn to dusk. The GVPP, on the other hand, adapts to real-time changes in lighting without recalibration, day or night.
Okay, so what they're claiming is that their brand-new, $6 ($50 in total) device can do things which long-standing scientific projects costing $10,000 cannot? Am I the only one who thinks that this sounds somewhat fishy?
The GVPP was invented in 1992, when BEV founder Patric Pirim saw it would be relatively simple for a CMOS chip to implement in hardware the separate contributions of temporal and spatial processing in the brain.
Again don't you think that all of the many computer scientists and neuropsychologists working on machine vision wouldn't have thought of this themselves? I've read a fair bit on the theory of vision processing and pattern recognition and it's a hugely complex subject. And now a small research company has cracked it? I don't think so. If you read the list of things which they say it can be used for it comes across as being a huge gimmick - they seem to have listed everything they could think of that might be worth money.
Actually it's not as impressive as it looks.
They say that The $6 Generic Visual Perception Processor (GVPP) can automatically detect objects and track their movement in real-time, according to Bureau d'Etudes Vision (BEV)
This can be easily accomplished by a technique called blob tracking - which is the coarsest image vision technique. A similar project (impressive too) was developed at a japanese company (although I don't remember exactly where). It was some sort of interactive game where a pet was playing with you in a projected image. You moved, played with it and it seemed to understant if you touched him,petted him etc. The catch was that the camera was filled with cameras and they detected the movement of your hands. Given the speed, direction etc. they could actually appreciate what you were supposed to do.
Cute, but nothing interesting from a research point of view.
I assume that they are doing the same. It is very easy to identify the movements in an image (you can do it in real time even on a Pentium). Check any image vision book for details.
Probably they built some chip that works at the speed of a controller (i.e. very fast) but, as any controller performs very few operations.
Still they don't say anything about actual image understanding.
And this is where comes the commercial part. Because they actually are not saying their chip can understand an image. They simply can track motion. That system wouldn't have a clue whether that movement is a fighter or a flying orange.
It may be useful in a computer that is used in a vision lab, but we're quite far from industrial pattern recognition or image understanding.
So please don't take commercial ads as truth
If you are French please don't read what will follow.
After all, they are French, and they are the best sellers in the world. Every one believes that French wines are great and French women are beautiful. Have you ever tasted those sour poisons ? Or ever went to France to watch their women ?
...if the general-purpose hardware wasn't so stupid. Of the millions of transistors on a modern chip, most of them are wasted in maintaining the illusion of sequential operation, while the OS writers go to considerable trouble to create the illusion of parallel operation.
Furthermore, there are the huge (in terms of transistor count) banks of flip-flops which just sit around most of the time, and the costly layers of cache all working their hardest to maintain the illusion that it is RAM. Meanwhile, software optimizers make sure to access memory sequentially to avoid upsetting this illusion, which would ruin the performance.
You can justify all this nonsense with the argument that software is written for sequential machines with RAM. It's a circular problem. If somebody would just release a cheap massively parallel system, the programmers would learn to use it efficiently.
You can make a complete processor in a few thousand transistors (as this guy has done, though he goes a bit off the deep end...), and you can add a bit (a few K) of high-speed RAM and network them easily enough to make a (dare I say it?) Beowulf cluster on a chip. Each might only run at one tenth the speed of a modern CPU, but you could have hundreds of them for the same cost, giving you bips and gflops for the price of mips.
It would also make the whole design process a lot easier and faster. One simple processor, repeated hundreds or thousands of times. Every advance in production would bring a direct and proportional improvement in performace, with a tiny added design cost. Forget special graphics or sound processors, just plug in more processor banks like you would add memory today and watch your system fly.
C'mon hardware guys, we software guys aren't that stupid! We don't need your illusion of a 386!
I'm sure they've done some great work but my personal belief is that it's been subject to a marketing department's hype machine. The vision system in a primate is one of the most complex parts of the brain and is still not entirely understood by neuropsychologists.
There are three main pathways from the eye - the magnocellular (connected to the rods mainly and used for brightness and motion detection), the parvocellular (connected to the cones mainly and used for colour determination) and the koniocellular (whose function is less well known). These three pathways feed into the V1 area which acts as a feature detector, then into the V2 area, which detects colour features and movement and then into a variety of different areas including the V3 (shapes), V4 (colour) and V5 (motion and positioning) areas, the parietal lobe, the thalamus and the various inferotemporal and interparietal areas among others still being found.
All of these different areas seem to have some bearing on vision in its entirety, and show just how complex vision is. As such I think that any claim that a company has suddenly perfected a chip which allows complex visual capabilities is suspect until hard facts and experimentation can prove or disprove the claims.
On the other hand, this could also be the basis for technology that tracks where you go and what you do. Under the auspices of controlling crime, criminals could be "flagged" and watched, traffic policing could be automated, etc. Where it gets scary is in who determines what suspicios behavior is, or who qualifies as needing to be watched, or the fact that you are removing the human element from the decision making process of evaluating a crime.
In the end, both the citizens and the govornment want this kind of pervasive, intelligent, monitoring technology to be ubiquitous. The difference is that citizens want to be able to turn it off.