NVIDIA has been hard at work on the problem posed by high frame rate interpolation of video data shot on lower fps. We have had this tech since the late 1990s with the advent of Twixtor and refined over the decades in systems like Twixtor Pro and Adobe’s Optical Flow in After Effects. You are still not getting real temporal detail data since the frames are created by extrapolating velocity and direction vectors plus pixel values between frames to get the result.
We explored this technique in our post on interpolation here and why it is no substitute from a real slow motion camera solution. NVIDIA’s new method uses machine learning along with 11,000 videos to arrive at a more convincing result. Considering the relatively small sample size we can imagine a future where hundreds of thousands or millions of footage samples are used to generate near flawless interpolation. This technique takes some serious computation and data sets so as of now it is not really ready for the mass market but that could change with the cloud very soon. → Continue Reading Full Post ←
The Rosalind Franklin Institute has started the development of a 100 million fps high-speed camera that will operate at 1-megapixel resolution to scan how new cancer drug treatments along with ultrasound interact to create effective cures for the deadliest forms of the disease. This type of performance is unheard of in that resolution and would herald a new way of looking at minute amounts of time that happen so fast no detector so far has been able to capture them at a usable resolution.
To put this in numbers, the camera will be able to capture a 100,000,000,000,000 or One Hundred Trillion Pixels per second or one hundred million megapixels/sec. Those are staggering numbers and if saved in a raw format it would take a megabyte per frame or 100 Terabytes of imaging data/second. That will be one extremely large frame buffer. With compression techniques and image optimization you could probably get that number lower but if played back at 30fps it would take 38 days to see a single 1-second video, staggering speed for sure! → Continue Reading Full Post ←
Sony is not only pushing the frame rate boundaries of CMOS sensors with their Stacked technology; but now it’s implementing and improving on the design with circuitry that can track objects with pinpoint electronic precision based on pixel colors, contrast, and motion vectors. The result the IMX382 is a complete tracking sensor solution that could one day be used for full AF during high-speed video capture by accurately representing the scene.
While 1000fps is not the end all be all of high-speed frame rate, the fact that it can accurately track with very little lag the object motion in front could one day translate into perfect autofocus capability for many types of cameras. Not only the focus department is improved by this technology but now you can use the sensor in all sorts of industrial, law enforcement, traffic cameras and monitoring solutions for robotics. → Continue Reading Full Post ←
Lytro started in the consumer space a few year’s back by enabling light field camera sensor technology in a portable package. Back then it allowed the user to select the focus point in the image to control depth of filed after the shot had been taken. This is emulated by Panasonic on the GX8 and GH4 with Post Focus but that is a feature that does some tricks with multiple images and lens focus points to select final depth of field.
Lytro does this by capturing all the rays of light entering the sensor at different angles and times to create a light field or three dimensional map of a real subject or scene. The consumer cameras sold in less than stellar numbers due in part to low resolution and while they tried with the Lytro Ilum to bring a more SLR like camera with 40 mega-rays or down-converted to 4 traditional megapixels. It also bombed as the post focus feature was not enough pull to get consumers to adopt the platform; along with claims of low dynamic range and artifacts in bright spots. The Ilum camera is still available for purchase at under $370 USD from the 1299 introductory price; quite a drop! → Continue Reading Full Post ←
The Latest on Hi Speed Affordable Imaging!
