Visage Technologies
Face Tracking & Analysis


September 8, 2015

Computational models for fashion models

Who’s going to be the next top model? Researchers from the Indiana University used Instagram as a dataset for their algorithm that can predict the popularity of new faces with an accuracy greater than 80%. They have used 400 models from the Fashion Model Directory, analyzed their Instagram accounts and their relevant statistics, for example […]

Continue reading →

September 3, 2015

From Skynet to robots on Mars: computer vision overview

  Curiosity saw the cat   Think about it: Mars is currently the only planet completely populated by robots! And after decades of studying machine learning, they are able to “see” the environment that surrounds them, move around freely while avoiding obstacles, and gather data using their sensory perception in order to find adequate samples. […]

Continue reading →
Faces in multimedia images usually exhibit different variations, dramatic expressions and illuminations

September 2, 2015

No more drama! Multimodal face-recognition techniques

Facial recognition algorithms nowadays have problems with extreme facial expressions, especially dramatic ones that appear in multimedia applications, social networks and digital entertainment. For example, various dramatic poses, different illumination or various expressions make things harder for computer algorithms to track and recognize faces. One of these niches with extreme facial gestures is sports – […]

Continue reading →
Top 10 results for matching Boston bombers in an 80-million galery

August 31, 2015

80-million gallery

Social media websites today are a vast gallery of photographs and pictures, and a great challenge for computer vision researchers is to devise algorithms and methods to index, search, track and identify millions of images and other media. That includes face tracking and detection on a large scale. IEEE researchers have devised a method which […]

Continue reading →
Neonatal (newborn) macaque imitating facial expressions

August 27, 2015

Learning to see by moving

Even though motor theories of perception have a long history, there’s has been a little amount of work in computational models of perception that make use of motor information. Nowadays the dominant paradigm in computer vision for learning various features relies on training neural networks to recognize objects using millions of annotated images. It seems […]

Continue reading →
← back  
Go to top