The Algorithm That Can Create Videos From A Single Photo

There are existing applications that turn a photo into a video. From Photoshop, to Premiere Pro, to After Effects, there are also plugins and other predictive algorithms that can do this task. This new algorithm however, aims to be the better version of the current applications for turning a photo to a video. Researchers at the University of Washington have developed a deep learning method that only needs a single photo to make a believable video, as DIY Photography details:  

Aleksander Hołyński explains that turning a photo into a video requires the algorithm to predict the future. “And in the real world, there are nearly infinite possibilities of what might happen next,” he adds. So, he and his team trained a neural network with thousands of videos of waterfalls, rivers, oceans, and other material with fluid motion. They would first ask the network to predict the motion of a video only by the first frame. Then, it would compare its prediction with the actual video, which helped it learn to identify clues that tell it what was going to happen next (such as ripples in a stream, for example).
The researchers tried to use “splatting,” a technique that moves each pixel according to its predicted motion. However, it posed another set of challenges. “Think about a flowing waterfall,” Hołyński told UW News. “If you just move the pixels down the waterfall, after a few frames of the video, you’ll have no pixels at the top!” So they had to come up with a solution for this, and they called it “symmetric splatting.” It doesn’t only predict the future, but also “the past” of an image, creating a seamless animation.

image credit: Sarah McQuate/University of Washington


Comments (2)

Newest 2
Newest 2 Comments

A single-processor computer can't multitask either -- just switch from one task to another, then back, very quickly. This may be applicable to human attention.

Though the human brain has many different parts which can operate simultaneously. I suppose a good definition of "multitasking" might be called for.
Abusive comment hidden. (Show it anyway.)
Evil Pundit - "This may be applicable to human attention."
And it is. One way that psychologists try to understand constructs that aren't directly measurable (like attention or memory) is to create a computer model - that is, how would a computer be able to achieve the same?
I know what you're thinking - this sounds stupid. How can you compare the biological brain to a mechanical computer?
But computer models have been fairly useful in helping us understand the possible ways that we for example, percieve, encode, and recall information.
Abusive comment hidden. (Show it anyway.)
Great:
This information make me feel little more relaxing now. This is probably the answer to much of the worries I 'm having some times. Good deal. So it's does make sense because it is like pushing the pedal to the metal. The more push the pedal the less efficient on combustion this will be like over flooding the brain in top of the whole things are already doing on auto-mode, Isn't? Good information way to go buddy.
Keep Cool...
MC
Abusive comment hidden. (Show it anyway.)
Login to comment.
Email This Post to a Friend
"The Algorithm That Can Create Videos From A Single Photo"

Separate multiple emails with a comma. Limit 5.

 

Success! Your email has been sent!

close window
X

This website uses cookies.

This website uses cookies to improve user experience. By using this website you consent to all cookies in accordance with our Privacy Policy.

I agree
 
Learn More