Abstract
Motion blur in an image is caused by the movement of the camera during exposure time; thus, awareness of the camera motion is a key factor in image deblurring algorithms. Among the various sensors that can be utilized while taking a picture in handheld devices, a gyroscope sensor, which measures the angular velocity, can help in estimating the camera motion. To achieve accurate and efficient single-image deblurring with a gyroscope sensor, we present a novel deep network with a flexible receptive field that is appropriate for training features related to the nature of the blur. Two specialized modules are sequentially placed in the proposed network to adaptively convert the shapes of the convolutional kernels. The first module directly transforms the kernel shape into the direction of the camera motion indicated by the gyroscope measurements. In the middle of the network, where the feature abstraction is sufficiently proceeded, the second module integrates features from the blurry image along with the information from the gyroscope to convert the kernel shape effectively, even when the gyroscope sensor is unreliable. Using a new gyro-image paired dataset, extensive experiments were conducted to show the effects of the reliability of the gyroscope measurements on the deblurring performance and to prove the effectiveness of our strategy.
Highlights
Despite the rapid development of handheld cameras, motion blur in an image is still visible when the device moves during the image exposure time
PROPOSED METHOD We present effective gyroscopeguided network (EggNet), which robustly transforms the receptive field by exploiting information from both the input image and the gyroscope for efficient image deblurring
In our previous work [33], we presented an approach that exploits a channel-wise stacking of the warped images in which blur artifacts caused by camera motion are aligned in the input data for the convolutional neural networks (CNNs) model
Summary
Despite the rapid development of handheld cameras, motion blur in an image is still visible when the device moves during the image exposure time. Motion blur affects the visual quality of the image and degrades the performance of various applications such as object detection, image segmentation, and visual odometry. Further improvements in single-image deblurring, in which a latent sharp image is recovered from a blurry one, are being actively researched. The non-blind methods [1], [2] employ a given blur kernel followed by deconvolution of the blurry image. Blind deblurring methods use only the blurry image to recover the sharp image. In many cases, blind methods [3]–[7] attempt to perform deblurring by first estimating the precise blur kernel from a blurry image and adopting the estimated kernel in a
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
