The google unblur is not the same as the Prophesee unblur patent I posted yesterday ( WO2021259994A1 IMAGE ENHANCEMENT METHOD, APPARATUS AND SYSTEM) because the patent requires simultaneous capture of the frame image and the DVS data.https://www.msn.com/en-au/news/tech...ogle-pixel-7-pro-e2-80-99s-camera/ar-AA12FVtc
Gizmodo Australia
FollowView Profile
We Zoom In on the Google Pixel 7 Pro’s Camera
Asha Barbaschow - 1h ago
During Made By Google this morning, the search giant unveiled its new flagship Pixel phones, giving us the rundown of what you can get when you mix Google software with Google hardware. The camera system on the Pixel 7 and 7 Pro phones don't appear to be too different from their predecessors, but when you zoom in, you see the new Google Tensor G2 chip working its magic.
As is always the case with Google, one super cool feature has piqued our interest in particular around the Pixel camera, and with the 7 Pro, it’s Photo Unblur.
The Google Pixel 7 and 7 Pro come packed with Google’s Tensor G2 chip, which powers all of the smarts behind the scenes. Such smarts include Face Unblur and now, Photo Unblur.
With the Pixel 6 and 6 Pro phones, Google introduced Face Unblur, which, as it says on the tin, brought the capability to unblur a person’s face in a photo after it was captured. Expanding on that tech this year, Google is using a new type of machine learning model to unblur more photos.
This has been added to Google Photos so you can unblur photos that have been taken in the past – they don’t even need to have been captured on a Google/Android device.
This runs on-device through a special optimisation on Google Tensor.
Just sayin'.
Late edition:
This is the Google unblur patent:
WO2022104180A1 SYSTEMS, APPARATUS, AND METHODS FOR REMOVING BLUR IN AN IMAGE
Systems, apparatus, and methods are presented for deblurring images. One method includes receiving an image and estimating blur for the image. The method also includes applying a deblurring filter to the image and reducing halo from the image.
[0033] The image processing system 202 may receive an input image. The input image may be processed by the pre-filter module 202 to reduce or remove noise from the image. For example, the pre-filter module 204 can denoise the input image to generate a noise-reduced image. The pre-filter module 204 may include a non-local means filtering to reduce or remove noise from the input image. In other embodiments, the pre-filter module 204 may use a denoising neural network (e.g., a convolutional neural network, or other suitable neural network) to remove or reduce the noise from the input image. In such cases, the denoising neural network can be trained by inputting multiple noisy versions of images together with clean versions of the same images. Using the known inputs (noisy images) and the known outputs (clean images), the denoising neural network can tune its parameters (e.g., weights, biases, etc.) to be able to output clean images (the noise-reduced images) from noisy images that are input into the neural network. The resulting noise-reduced images can have no noise or may still have some noise, but the noise may be greatly reduced by the denoising neural network.
Last edited:
’s
