Robots that recognize when people are stupid

Lost the remote? Use your face | NetworkWorld.com Community

By using a combination of facial expression recognition software and automated tutoring technology Jacob Whitehill, a computer science Ph.D. student from UC San Diego’s Jacobs School of Engineering, is leading the project that ultimately is part of a larger venture to use automated facial expression recognition to make robots more effective teachers.

The researchers recently conducted a pilot test with 8 people that demonstrated information within the facial expressions people make while watching recorded video lectures can be used to predict a person’s preferred viewing speed of the video and how difficult a person perceives the lecture at each moment in time.

“If I am a student dealing with a robot teacher and I am completely puzzled and yet the robot keeps presenting new material, that’s not going to be very useful to me. If, instead, the robot stops and says, ‘Oh, maybe you’re confused,’ and I say, ‘Yes, thank you for stopping,’ that’s really good,” said Whitehill in a release.

Is there anything people can’t do with a little bit of image processing?

Because the Times has not yet discovered the 3d dimension

Scientists find bugs that eat waste and excrete petrol - Times Online

The closest that LS9 has come to mass production is a 1,000-litre fermenting machine, which looks like a large stainless-steel jar, next to a wardrobe-sized computer connected by a tangle of cables and tubes. It has not yet been plugged in. The machine produces the equivalent of one barrel a week and takes up 40 sq ft of floor space.

However, to substitute America’s weekly oil consumption of 143 million barrels, you would need a facility that covered about 205 square miles, an area roughly the size of Chicago.

Obviously if you can feed the things on something you don’t want (i.e. not corn or any other food) you’re in good shape — roughly as good as with cellulosic ethanol.

Back of the envelope says 143 billion liters is 143 million cubic meters, which is 143 meters high by a kilometer on a side. That sounds rather smaller (depending on how it’s divvied up) that the volume occupied by refineries at the moment. Of course, you’d also need to be moving roughly that volume of biomass around…

Send the colony ships now

BBC NEWS | Science/Nature | Huge ice deposits ’seen’ on Mars

The ice is found in distinctive geological structures on Mars’ surface that are hundreds of metres thick.
….
Mission scientists used Sharad to probe Martian surface features known as lobate debris aprons (LDAs). These distinctive, dome-shaped structures are concentrated around mid-latitudes in the planet’s northern and southern hemispheres. Geological clues The researchers looked at LDAs in the Deuteronilus Mensae region of Mars’ northern hemisphere, where the features can be found at the bases of valley walls, craters and scarps of mesas. Scientists have long suspected that LDAs were flows consisting of mixed up rock and ice. The radar penetrated these geological features with very little attenuation (reduction in signal strength), suggesting they were predominantly made of ice. “We would say, robustly, more than 50% ice by volume - but it could be much more,” said Jeff Plaut, the chief scientist for Sharad.

A quick look at some of the pictures suggests that we’re talking about flows a few kilometers across. At 50% ice that would be more than a cubic kilometer of water per flow. That’s a billion tons of water, aka somewhere north of 200 billion gallons. You could support Las Vegas for days with that kind of water.

Scavenging once again

Images: This knee brace could charge batteries | CNET News.com:

Brace yourself for the latest in battery-charging technology for gadgets. Researchers from Simon Fraser University in British Columbia have developed a device worn much like a knee brace that generates electricity from the natural motion of walking

I want Eliza to sign to me

BBC NEWS | Technology | IBM animates sign language avatar:

SiSi will enable deaf people to have simultaneous sign language interpretations of meetings and presentations. It uses speech recognition to animate a digital character or avatar. IBM says its technology will allow for interpretation in situations where a human interpreter is not available. It could also be used to provide automatic signing for television, radio and telephone calls.

Lucky Imaging

Institute of Astronomy:

The camera works by recording the images produced by an adaptive optics front-end at high speed (20 frames per second or more). Software then checks each one to pick the sharpest ones. Many are still quite significantly smeared but a good percentage are unaffected. These are combined to produce the image that astronomers want.

This is really quite cool, although I worry a bit about artifacts (is it automatically true that the “sharpest” images are the most accurate representations?). I’m also a little awed by the advances the system takes for granted: adaptive optics, CCD sensors big and efficient enough to capture useful number of photons in 50 milliseconds…