Today we announced a new feature on Pixel 7/Pro and @GooglePhotos called "Unblur". It's the culmination of a year of intense work by our amazing teams. Here's a short thread about it

1/n

#MadeByGoogle #fixedonpixel #Pixel7 #PhotoUnblur
@GooglePixel_US

Last yr we brought two new editor functions to Google Photos: Denoise & Sharpen. These could improve the quality of most images that are mildly degraded. With Photo Unblur we raise the stakes in 2 ways:

First, we address blur & noise together w/ a single touch of a button.

2/n
Second, we're addressing much more challenging scenarios where degradations are not so mild. For any photo, new or old, captured on any camera, Photo Unblur identifies and removes significant motion blur, noise, compression artifacts, and mild out-of-focus blur.

3/n
Photo Unblur works to improve the quality of the 𝘄𝗵𝗼𝗹𝗲 photo. And if faces are present in the photo, we make additional, more specific, improvements to faces on top of the whole-image enhancement.

4/n
One of the fun things about Photo Unblur is that you can go back to your older pictures that may have been captured on legacy cameras or older mobile devices, or even scanned from film, and bring them back to life.

5/n
It's also fun to go way back in time (like the 70s and 80s !) and enhance some iconic images like these photos of pioneering computer scientist Margaret Hamilton, and basketball legend Bill Russell.

6/n
Recovery from blur & noise is a complex & long-standing problem in computational imaging. With Photo Unblur, we're bringing a practical, easy-to-use solution to a challenging technical problem; right to the palm of your hand
w/ @2ptmvd @navinsarmaphoto @sebarod & many others

n/n
Bonus: Once you have a picture enhanced with #PhotoUnblur, applying other effects on top can have an even more dramatic effect. For instance, here I've also blurred the background and tweaked some color and contrast.

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Peyman 𝕄𝕀𝕃𝔸ℕ𝔽𝔸ℝ

Peyman 𝕄𝕀𝕃𝔸ℕ𝔽𝔸ℝ Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @docmilanfar

Oct 8
Yesterday @ #MadeByGoogle we also announced the latest on the Super Res Zoom feature on #Pixel7Pro

It's a project my team's been involved in since '18. This year, our teams've made it so much more powerful. You can zoom up to 30x. Let me show you in a 🧵

Image
That👆one was the ultra-wide lens (0.5x).

This 👇one starts at 1x and get closer.... ImageImage
And closer.... ImageImage
Read 6 tweets
Aug 10
Here's a neat trick to impress your friends
Let's say you have some curve with a random shape, possibly even self-intersecting. Can you measure its length?

This isn't just a parlor trick -- it has many practical applications. For example, the curve could be a strand of DNA
1/n
You can think of the curve as a collection of tiny segments of course. You can then measure each segment and add up the results.

OK, but you can go further and take the segments to be so small that they are almost like points. You can then add up the (red) "points".

2/n
In practice, this is not easier though. So we need a mechanism for doing this more conveniently. One way to do it is to drop lines that intersect with the shape and count the number of intersections. You can do this with a mesh too.

3/n
Read 7 tweets
Jul 12
This is a story about a guy you might have heard of.

A guy who understood that our journeys here on earth are short. That we’re all passengers pitching downward into the night

Yet he gave us stories with hope; so we can get up and move forward, explore, & help each other

1/10
2/10
3/10
Read 11 tweets
May 22
I've posted a couple of dozen technical threads in the last two years that have been popular. Here they are in a thread of threads, very roughly organized into related themes.

Let's start w/ little-known classic: mean & median within a standard deviation

Stein's lemma and characterization of the Gaussian

Read 23 tweets
May 15
I grew up eating some amazing sandwiches like the Kalbas (beef Mortadella) & Sosis (fried sausage) in Tehran. These are a meal and a half.

There's a crazy variety of sandwiches around the world. I've come across some memorable ones🧵:

1/n
Medianoche (Cuba):

Roast pork, ham, mustard, Swiss cheese, and pickles served on sweet bread. It is so named because of the sandwich's popularity as a staple served in Havana's night clubs right around or after midnight.

2/n
Chip Butty (UK):

Filled with chips (thick-cut deep fried potato, not to be confused with thin-cut french fries), optionally eaten with condiments such as brown sauce, ketchup, mayonnaise, or malt vinegar.

3/n
Read 13 tweets
Mar 14
Sequential estimators (eg Kalman Filter) are hard to grok especially when couched in vector language.

For intuition, one can start from very simple: consider just least-squares on a scalar A.

Say we have n=0:N-1 noisy (0-mean & iid) measurements of A

x(n) = A + w(n)

1/7🧵
The least-squares estimate of A is simply the average of the N data points x(0), ... , x(N-1) . Now suppose a new data point x(N) becomes available

We can find the LS estimate all over, but that's wasteful for both memory and compute. Much more efficiently, we can update:

2/7
It's neat to observe that in this simple case, the update equation shows that the "correction" term should go to zero as N→∞

The situation is quite similar if the noise variance σₙ² is different from sample to sample. We can still update the estimate just as well

3/7
Read 8 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

:(