OMG! She kissed him!

The people at the University of Oxford have developed a computer vision based system that detects interactions between people. The system can recognize four gestures as of now: Hand shakes, hugs, kisses and high-fives.

It uses an upper body detector and then uses cues like the orientation of the head, relative distance between people, etc

The Active Vision Group and the Visual Geometry group

Matlab now supports GPGPU

The latest version of Matlab now supports General Purpose GPU. The Parallel Computing toolbox comes with support for nVidia CUDA devices. [updates]

Self-recharging robots

Several companies are working on taking recharging to the next level - Robots that locate wall sockets and plug themselves in. The robot has several sensors and an arm that locate the socket and plug into the socket.

This got me wondering, can there be a machine-vision-only solution to this problem? A socket should be a very easily distinguishable thing. Hmm... could the Scale Invariant Feature Transform do this trick? Or will simple template matching work?

[Source]

OpenCV + RC Helicopters

Found an interesting thread. This person wants to make a remote controller helicopter with a camera + processor on it. And then do computer vision using OpenCV. Sounds like something Google should be researching about!

OpenCV + RC Helicopters

Found an interesting thread. This person wants to make a remote controller helicopter with a camera + processor on it. And then do computer vision using OpenCV. Sounds like something Google should be researching about!

RoboRealm's AVM Navigator

RoboRealm has decided to include a 3rd party plugin called AVM Navigator. This module lets you track objects or navigate to it. It operates in three modes:

• Object detection: Train it, and the module will recognize objects
• Navigation mode: Guide a robot to a recognized object
• Nova Gate mode: Guide a robot to a recognized object using several objects in-between.

[AVM Navigator on RoboRealm]

Humanoid learns archery with visual feedback

A humanoid baby bot has learned how to aim arrows. The Italian Institute of Technology has used the ARCHER (?!) algorithm to train the robot to learn archery on its own, using visual feedback. It needs 8 shots to fully learn how to shoot perfectly from a distance of around 11 feet.

PS: ARCHER stands for Augmented Reward Chained Regression
PPS: Time to get started with computer vision!

[Source]

Basic humanoid walk tutorial

I found a step by step tutorial for making a humanoid walk. The video uses the bioloid premium kit to take you through some of the basic ideas of designing a gait for your robot.

Google's self-driving cars

Google has been secretly working on probably one of the most important applications of artificial intelligence - self driven cars. Their cars use cameras, radar sensors and laser range finders to "see". And guess what, they've teamed up with the winners of the DARPA challenge.

[Source]

Robots ask 'What is life?" and synthetic bacteria

Humans are nothing but machines. Our "sadness" has a number associated with it... biochemicals. Robots can solve problems, and robots can behave intelligently. Robots' "sadness" is also just a number. Why should their "sadness" be lesser, say, when you turn them off? In fact, humans have created a synthetic bacteria. And it's "alive" alright.

[source]