This is the world’s first pair of glasses which prevent facial recognition by cameras. They are currently under development by Japan’s National Institute of Informatics.

Photos taken without people’s knowledge can violate privacy. For example, photos may be posted online, along with metadata including the time and location. But by wearing this device, you can stop your privacy from being infringed in such ways.

“You can try wearing sunglasses. But sunglasses alone can’t prevent face detection. Because face detection uses features like the eyes and nose, it’s hard to prevent just by concealing your eyes. This is the privacy visor I have developed, which uses 11 near-infrared LEDs. I’m switching it on now. It prevents face detection, like this.”

“Light from these near-infrared LEDs can’t be seen by the human eye, but when it passes through a camera’s imaging device, it appears bright. The LEDs are installed in these locations because, a feature of face detection is, the eyes and part of the nose appear dark, while another part of the nose appears bright. So, by placing light sources mostly near dark parts of the face, we’ve succeeded in canceling face detection characteristics, making face detection fail.”

Compared with previous ways of physically hiding the face, this technology can protect privacy without obstructing communication, as all users need to do is wear a pair of glasses.

However, because this system utilizes the difference in spectral sensitivity between human vision and imaging devices, another method is needed for cameras that aren’t affected by infrared light.

“In that regard, what we’re thinking of is a visor that doesn’t use electricity, but uses reflective material. For example, one like this. This makes light from outside look white, or absorbs it. That pattern breaks up the features used in face detection. So you can prevent face detection even without using electricity, by wearing this visor. It is also very cheap to make.”

It’s geriatric Big Brother!

The system was inspired by the person-tracking Marauders’ Map featured in Harry Potter books, and it’s called multi-camera or multi-object tracking. Previous attempts at multi-object tracking have had limited success, accurate either one third or one half the time. But those systems were tested in tightly controlled labs. Carnegie Mellon decided to try a more organic environment. A nursing home is a great testing environment, the researchers say: cameras already exist and have to deal with realistic obstacles like inconvenient furniture placement, doors getting in the way, blind spots, and residents moving freely. It’s also good because a tracking system in a nursing home reads as altruistic – it’s important to be able to find and care for the elderly as soon as they might need assistance.

The technology works through a combination of facial recognition and color tracking. Colored clothing is a good way to identify people, because it is visible most of the time, but the same color shirt can look different under different lighting. Algorithms compensate for differences in color appearance under different light–they make it so that you can track someone wearing a red shirt as he moves from a dark hall to a brightly lit dining room. Facial recognition is the best way to identify people, but faces are rarely pointed directly at cameras, so it only works about 10 percent of the time.

That’s why it’s important to track both faces and colors at the same time. The process resembles how cell phones pinpoint personal location with different inputs. Signals sent to cell towers provide a constant, rough idea of where the phone user is, and occasionally a GPS double-checks the position and corrects it if need be.

Carnegie Mellon’s program isn’t yet ready for prime time. The researchers used 6 minutes of footage recorded by 15 cameras in a nursing home in 2005 to develop the algorithms and test the system. A live trial is still a long way off, but once the system completes identification in a real-time setting, expect it to move from nursing homes to prisons to casinos and then everywhere.

People wanting to say hidden from this might just turn to facial-recognition-thwarting makeup.

[Kurzweilai.net]

Be like the bat. Echolocate.

The system could fairly accurately be described as echolocation, just like bats use: it measures the time it takes for a sound to produce an echo at different points in the room. Essentially, what they’ve come up with is an array of microphones and an algorithm that picks up both the original source and its echoes. Our ears can’t hear the tiny lags that make up the echoes in most sounds, but bats can, and so can this system.

From a single sound, they can reconstruct a room to within a few millimeters–provided the room isn’t too complicated, at least for now. The system outputs a 3-D map, which until now could only be made with visual tools like LIDAR.

What could this be used for? Well, there are virtual reality possibilities–the Kinect, for example, might someday use audio as well as visual clues to more accurately map action. Or it could be used in forensics. A simple audio recording could reveal the shape of a room in which a crime was committed–a valuable clue to which we wouldn’t have access before this.

The paper is published in the current issue of Proceedings of the National Academy of Sciences.

[via BBC]

“The Chime” is loaded with sensors for picking up ambient noise, then transforming that noise into music.

A city doesn’t always sound great: car horns blare, people shout. But a new project from artist Marc De Pape makes music out of the noise.

“The Chime” is a wind-chime-shaped mass of sensors that pick up on the environment: if the sensors detecting something approaching, the device tinkles with xylophone notes; if something is moving away, it swells with the sound of strings; and the music all shifts key based on the temperature. The result is a changing, ambient soundtrack for a city. De Pape explains the project like this:

Inspired by Georg Simmel’s notion of the Blasé (an indifference towards the difference between things), I set out to explore the relationship between sensing technology and the routines of everyday life. I feel the city is all too commonly represented by abstract systems and maps, a tendency driven by a reductionist pursuit of efficiency, and one which ignores the idiosyncrasies occurring on street level. This is the noise in the system, the richness that ultimately renders cities generative landscapes. I thus set out to bring attention to the noise by building a musical instrument inspired by wind chimes: The Chime is a collection of 18 sensors measuring 27 parameters assembled to poetically translate the impulses and flows of the everyday city into sound.

You can check more videos of the project over at De Pape‘s site. It’s much nicer to hear than people talking on their phones.

[Marc De Pape via Creative Applications Network]

And you thought the U.S. had deserts!

As beautiful as the maps turned out, Minar created the project largely just as a tutorial on how to make a vector-based map. (His code and more background on the process are on GitHub.) “It’s mostly a demo project with readable source,” he writes on his blog, “but it’s also kind of pretty.”

This particular map includes all flow lines, which is why you see a lot more blue than you might expect in desert areas–it encompasses seasonal water flow, like creek beds that are dry for much of the year. Down in Florida, the Everglades don’t have well-defined enough flow lines, so the swampy preserve isn’t included, and the state looks mysteriously white in comparison to, say, New Mexico.

And here’s California and some of the surrounding area (look how all the squiggles converge on the California Delta near the San Francisco Bay over at the left):

You can see the full zoomable map and pinpoint precise waterways here.

[Gear Junkie]