Thinks

iPad measures heart and breathing rates using only camera.

Philips Vital Signs Camera app measures fluctuations in skin colour and movement of shoulders to estimate your heart rate and breathing rate. It is impressive that skin colour can be tracked so accurately and quickly. Smartphone and tablet cameras are usually of poor quality (particularly forward facing cameras) so it is a testament to the algorithms that the system appears to work so well. See the review link below where they put it through some paces.

Integrate some of this technology into video conferencing and your tablet will be able to tell you when the person you are talking to is lying or not, or whether you’ve just embarrassed them. This would be useful to add to the extension of SixthSense that designer Timothy Byrne of Western Washington University is working on to provide social cues to suffers of autism.

(via Technology News: Reviews: Philips iPad App Uses Camera to Monitor Your Vitals With Surprising Accuracy)

SpeckleSense is a fast (1000 fps), precise (50 μm), low-cost (cheap components and no lenses) and compact motion sensor that exploits ‘Laser Speckle’.

Ever looked at a laser dot and seen a speckle pattern that seems to shift about as you move your head? That’s Laser Speckle. Researchers from Camera Culture Lab have exploited this by re-purposing the tracking sensor from an optical mouse to track the speckle pattern to sense motion in 3D, see the paper for more details. 

It’s compact and cheap which means you could have these built into mobile phones extremely cheaply, providing 3D mouse input control for mobile devices. 

It has similarities to Kinect technology, i.e. using an optical sensor to analyse a laser pattern project from a similar place. So Kinect on your phone might not be that far away.

(via SpeckleSense: Fast, Precise, Low-cost and Compact Motion Sensing using Laser Speckle)

I was reminded about Bokodes from Camera Culture Lab at an AR talk I gave recently. The system uses a 3mm microdot of tiled fiducial markers and a tiny lens that enables a portion of the markers to be read when the camera lens is focused on infinity.

Pros:

• Encode a large amount of information into a tiny space
• Pose information can be encoded into the matrix to assist augmented reality
• Active and passive types
• Do not have to be too close to the item in question

Cons

• Two cameras required if you want to do AR
• Non-trivial to create (unlike printed codes)
• Decent optics required for zoom and wide aperture, maybe?

The current process looks too expensive to be used mass market but maybe a holographic implementation could be the way to exploit this phenomenon much cheaper (by mitmedialab)

[Update] It looks like a a similar technology has been developed in a collaboration between Max Planck Institute for Informatics and MIT Media Lab, Cambridge, Massachusetts a few years. By using an array of these lenses (each one presenting a pixel), hundreds of photos (same object, different views) are compressed into one that changes depending on viewpoint. Here is the article with a video example of a photo with shifting shadow.

Immersive video puts you where the camera is.

A bit like still quicktime panoramic photos, the neat use of fish-eye lenses and iPad lets the viewer pan around as the film plays; not dissimilar to an on-rails 3D computer game. Looks like it would be most immersive when using the accelerometer to physically look around.

(via New iPad App Puts Viewers Inside Immersive Video | Raw File)

Interesting proof of concept for ubiquitous AR.

Obviously, we’re not going to be really using pizza boxes as laptops. That’s just silly. And I’m not sure about needing to kit every room out with the technology - I think the more personal approach that Pattie Maes et al are following with SixthSense has the potential for a more ubiquitous and personal augmentation of the world for users.

Nevertheless, some of the technology and concepts demonstrated here relating to supplementing real world objects with artificial affordances would go well with SixthSense. 

(via Turn Pizza Boxes Into Computer Interfaces With “Invoked Computing” (Video) | TechCrunch)

Tesco are the first major supermarket in UK to trial AR so that customers can see how big products might be before ordering them. The web cam-based AR developed by KishinoAR works well for toys and items you can hold in your hands but I think furniture items like the TVs (fast forward to 1:35 in the video) are crying out for a mobile AR implementation so you can try them in their intended locations around the house.

(via KishinoAR’s Channel - YouTube)