|Back to Jue Wang's homepage|
Image and video Deblurring
last modified: 7/31/2013
Taking handheld photos in low-light conditions is challenging. Since less light is available, longer exposure times are needed – and without a tripod, camera shake is likely to happen and produce blurry pictures. Increasing the camera light sensitivity, i.e., using a higher ISO setting, can reduce the exposure time, which helps. But it comes at the cost of higher noise levels. Further, this is often not enough, and exposure times remain too long for handheld photography, and many photos end up being blurry and noisy.
We aim to develop practical solutions to restore a blurry input image caused by camera shake, and produce a sharp, or enhanced version of it. Such solutions have many real-world applications, from photo enhancment to image forensics. Most importantly, we want our tools to be fast and user-friendly, and to be robust against other artifacts that are common in real photographs: noise, JPEG compression artifacts, and non-linear tone mapping.
Below is an image motion deblurring example, you can find more examples here.
a blurry input image enhanced by image deblurring
Beside camera shake, a photography can also be blurry even the camera is mounted on a tripod and the scene is in focus. This is because DLSR lens all have the sharpness fall-off problem near the corner of the images. We also want to correct this type of blur by lens calibration.
Photoshop CC, Shake Reduction, 5/2013
After years of research, we sccessfully built the Adobe image deblur engine and transferred it into Photoshop CC in May 2013, under the product name "Shake Reduction". Some of the advanced features of this tool include:
Below is a list of research papers we published on this sbuject:
Camera shakes happen more often with a video camera. Significant camera shake will cause video frames to be blurry. Restoring shaky videos not only requires smoothing the camera motion and stabilizing the content, but also demands removing blur from video frames. However, video blur is hard to remove using existing single or multiple image deblurring techniques, as the blur kernel is
both spatially and temporally varying. Below is an example.
original video frame deblurred video frame
We aim at developing a practical (meaning that we can acutally ship it in a product!) video deblurring method that can effectively restore sharp frames from blurry ones caused by camera shake. Our method is built upon the observation that due to the nature of camera shakes, not all video frames are equally blurry. The same object may appear sharp on some frames while blurry on others. Our method detects sharp regions in the video, and uses them to restore blurry regions of the same content in nearby frames.