Image Deblurring using Inertial Sensors

[Gavin Greenwalt]

http://research.microsoft.com/en-us/um/redmond/groups/ivm/imudeblurring/

We present a deblurring algorithm that uses a hardware attachment coupled with a natural image prior to deblur images from consumer cameras. Our approach uses a combination of inexpensive gyroscopes and accelerometers in an energy optimization framework to estimate a blur function from the camera’s acceleration and angular velocity during an exposure. We solve for the camera motion at a high sampling rate during an exposure and infer the latent image using a joint optimization. Our method is completely automatic, handles per-pixel, spatially-varying blur, and out-performs the current leading image-based methods.

Interesting research that RED could apply with existing camera features.

 

[Curran Giddens]

iiiinteresting....

 

[Stephen Strangways]

They are sampling over time, so this would only be useful for stills mode, not motion.

 

[Mohammed El Sharqawy]

I think RED has integrated their stabilization system similar to this.. it puts the stabilization as a metadata with the recorded files..
right it would be nice for RED one to have stabilization module to be added on..

 

[Gavin Greenwalt]

They are sampling over time, so this would only be useful for stills mode, not motion.

There's nothing stopping them from increasing to sub-frame inertial sampling. I imagine they just have a sub-routine averaging the sub-frame input anyway.

Certainly inferior to an optical image stabilizer but if you bundled in a 100KB gyro/accelerometer stream from each frame you might want to experiment with the effect, just like you experiment with noise reduction now.