Barcelona, 19 de marzo del 2014.- The facial recognition software is closer than ever 100% and the miniaturization of devices reaches incredible action. In Forbes magazine I’ve read the latest developments in your Facebook facial recognition:
«Have you noticed that Facebook is getting better at making suggestions for people to tag in the photos that you have uploaded? Facebook will only get better at identifying faces thanks to advances in artificial intelligence and “deep learning.” Facebook researchers are currently developing algorithms called “DeepFace” to detect whether two faces in unfamiliar photos are of the same person with 97.25% accuracy, regardless of lighting conditions or angles. As a comparison, humans generally have an average of 97.53% accuracy. This means that Facebook’s facial-processing software has nearly the same accuracy as humans.
Yaniv Taigman, one of Facebook’s artificial intelligence scientists, said that the error rate has been reduced by over 25% relative to earlier software that handles the same task. Taigman co-founded Face.com in 2007, which was acquired by Facebook a couple of years ago. Prior to the acquisition, Face.com built apps and APIs that could scan billions of photos every month and tag faces in those photos. Taigman developed DeepFace with fellow Facebook research scientists Ming Yang and Marc’Aurelio Ranzato, along with Tel Aviv University faculty member Lior Wolf.
According to MIT’s TechnologyReview.com, DeepFace uses a 3D model for rotating faces virtually so that the person in the photo appears to be looking at the camera. The angle of the face is corrected by using a 3D model of an “average” forward-looking face. DeepFace creates a simulated neural network to work out a numerical description of the reoriented face to determine if there are similar enough descriptions from the two images. This network involves over 120 million parameters using locally connected layers. The DeepFace team trained the network using a dataset of 4 million facial images belonging to around 4,000 people, which means that each identity had an average of over a thousand samples for testing. The facial verification technique can be used to complement the connection of a name to a face, known as facial recognition. Eventually, this could improve Facebook’s ability to suggest users for tagging in an uploaded photo and for other potential purposes. The DeepFace algorithms have also been successfully tested for facial verification within YouTube videos, but this was challenging because the imagery was not as sharp compared to photos»
Wired Magazine: «You Can Take Selfies of Your Aorta With This Mini Camera»
Another technological innovation that has surprised me is the following article in Wired magazine. I reproduce below a summary, the full report here
Scientific studies of selfies have yielded interesting insights on personalities, gender differences, and national moods, but scientist F. Levent Degertekin has invented a new camera that can provide high-def, 3-D images of your innards. This “camera” uses ultrasound imaging techniques to create real-time, volumetric images of occlusions in arteries, but it’s built more like a miniature drum cymbal than a SLR. A donut-shaped silicon chip with a 1.5 millimeter diameter and 460 micron hole in the center houses sensing and transmitting circuitry and serves as the base of the diminutive device. A thin film on top of it flutters 0.00005 of a millimeter, creating sound waves which are captured by an array of 100 sensors on the chip, processed, and transmitted to an external video monitor at a rate of 60 frames per second via 13 gossamer cables that are threaded through a catheter.The images it produces are able to replace two people in the surgical theater. Though it’s roughly the size of a grain of uncooked quinoa, the images it produces are able to replace two people in the surgical theater. Prior to the invention of this speck-sized sensor, technicians would pore over lower-fidelity cross-sectional images and guide the surgeon verbally while she held the patient’s life in her hands. Degertekin likens his little invention to a flashlight that illuminates the obstructions in a blood vessel, giving doctors a direct look at what they’re up against.
The tiny camera is a breakthrough a decade in the making. Creating imagery with sound waves is old news, but being able to do so in a circulating blood vessel without killing the patient is a tough balancing act. Surgeons demand high-fidelity images, but high-power tools often heat up, causing the delicate silicon elements to fail and the patient’s blood to boil. Clever power controls allow Degertekin’s system to satisfy both audiences.