Computer Vision Blog
An inspiring talk by David Lang co-founder of OpenROV, a community of ocean lovers who build underwater robots, which shows how opensource technology can expedite ocean exploration.
The Department of Homeland Security is working assiduously to develop a surveillance system integrated with a crowd facial scanning technology. Biometric Optical Surveillance System (BOSS) is an unfinished version of the facial scanning technology that is now being developed.
There already exist computer vision systems that are able to perform a few tasks associated with the human visual system such as identifying faces or adjusting the incoming light, but we are still far from a system that has the ability to search for and identify specific objects, in the way that the human visual system would.
Since the early 1950s humans have been plagued with promise that all jobs and tasks would eventually be taken over by a machine. Well, thanks to the modern computer and increasing processor speeds, we have never been closer to this eventuality.
To avoid the delays associated with transferring data between Earth and Mars, NASA scientists have developed software that allows the rovers roaming Mars to analyze on the fly. Developing vision software for use in space applications has proven to be simpler than developing similar software for use on our home planet.
Students at Stanford University are taking photos of birds' movements that were previously invisible to the human eye. The photos are being used to study everything from the movements the winged-creatures make just before taking off and landing to the length and position of their legs during flight. The information is being studied as it relates to the development of UAV.
Move over Lytro, Focii is entering the 3D scene. Researchers in the Camera Culture Group at MIT’s Media Lab have developed a system that can produce multi-perspective 3D images in addition to enabling images to be refocused post-capture, all from a single exposure.
A team of scientists has been working on a “Coral-bot” that could go underwater and repair the earth’s dying reefs. Reefs have been damaged through both increasing human pollution, and as always by natural forces as well. The importance of up-keeping the well-being of these reefs cannot be overstated.
A recent system based off of Kinect technology has been developed to detect depression with an impressive 90% accuracy rate. This development goes back to the notion that the essential groundwork for computer vision breakthroughs is readily available to us, and that we just need to know about it to continue advancing forward in this field.
Machines on assembly lines are no revelation; they’ve been around for years. Often times it is more efficient to have a robot perform the same repetitive task several thousands of times over. Because of the nature of these assembly line-robots, they often have to be in a separate workspace from humans.