ADAPTIVE MESH-GRID BASED SPATIALLY NON-UNIFORM VIDEO MOTION DEBLURRING USING IMU-CAMERA SENSOR FUSION

2021-9-6
Arslan, Ahmet
Motion blur is a common problem that produces blurry images with inevitable information loss and unpleasant visual quality. Relative motion between scene and image recording device during exposure time causes the motion blur. This artifact degrades the human perceptual quality and also challenges subsequent computer vision applications. Most of the studies in the literature assume uniform blur present in the image and focuses on removing motion blur with non-blind deconvolution algorithms because of their low computational cost. Low-light conditions and rotation around the optical axis violate this uniform motion blur assumption and decrease the deblurring performances. In this thesis, we developed a novel adaptive mesh-grid based motion deblurring algorithm to handle non-uniform motion blur and decrease the high computation cost of existing methods. We use an inertial measurement unit (IMU) for obtaining the motion during the exposure period and generate a point spread function (PSF). The mesh-grid implementation is simply dividing an image into sub-block and performing blur kernel estimation and deblurring separately for each block. In addition, a variance parameter is proposed for representing the level of blur non-uniformity. To increase PSF estimation accuracy, mesh-size changed according to this variance parameter. Using this approach, two adaptive mesh-size algorithms are developed for the best quality and optimal quality with low computation time. It is observed that the proposed adaptive mesh-size algorithm increases the performance and quality of the existing methods.

Suggestions

Multi-Frame motion deblurring of video using the natural oscillatory motion of dexterous legged robots
Gultekin, Gokhan Koray; Saranlı, Afşar (2019-07-18)
Motion blur is a common problem for machine vision applications in legged mobile robotic platforms. Due to the oscillatory walking nature of these platforms, the cameras on them experience disturbances so that most of the captured frames are motion blurred. Motion blur results in loss of information in individual image frames and therefore makes single-frame deblurring an ill-posed problem. The variation in the magnitude of motion blur in consecutive video frames can be exploited to better restore these fra...
FRAME-RATE CONVERSION FOR MULTIVIEW VIDEO EXPLOITING 3D MOTION MODELS
Gedik, O. Serdar; Alatan, Abdullah Aydın (2010-09-29)
A frame-rate conversion (FRC) scheme for increasing the frame-rate of multiview video for reduction of motion blur in hold-type displays is proposed. In order to obtain high quality inter-frames, the proposed method utilizes 3D motion models relying on the 3D scene information extractable from multiview video. First of all, independently moving objects (IMOs) are segmented by using a depth-based object segmentation method. Then, interest points on IMOs are obtained via scale invariant feature transform (SIF...
Color appearance in high-dynamic-range imaging
Akyüz, Ahmet Oğuz (2006-07-01)
When viewing images on a monitor, we are adapted to the lighting conditions of our viewing environment as well as the monitor itself, which can be very different from the lighting conditions in which the images were taken. As a result, our perception of these photographs depends directly on the environment in which they are displayed. For high-dynamic-range images, the disconnect in the perception of scene and viewing environments is potentially much larger than in conventional film and photography. To prep...
MULTI-RESOLUTION MOTION ESTIMATION FOR MOTION COMPENSATED FRAME INTERPOLATION
Guenyel, Bertan; Alatan, Abdullah Aydın (2010-09-29)
A multi-resolution motion estimation scheme is proposed for tracking of the true 2D motion in video sequences for motion compensated image interpolation. The proposed algorithm utilizes frames with different resolutions and adaptive block dimensions for efficient representation of motion. Firstly, motion vectors for each block are assigned as a result of predictive search in each pass. Then, the outlier motion vectors are detected and corrected at the end of each pass. Simulation results with respect to dif...
Correlation tracking based on wavelet domain information
Ipek, HL; Yilmaz, I; Yardimci, YC; Cetin, AE (2003-08-07)
Tracking moving objects in video can be carried out by correlating a template containing object pixels with pixels of the current frame. This approach may produce erroneous results under noise. We determine a set of significant pixels on the object by analyzing the wavelet transform of the template and correlate only these pixels with the current frame to determine the next position of the object. These significant pixels are easily trackable features of the image and increase the performance of the tracker.
Citation Formats
A. Arslan, “ADAPTIVE MESH-GRID BASED SPATIALLY NON-UNIFORM VIDEO MOTION DEBLURRING USING IMU-CAMERA SENSOR FUSION,” M.S. - Master of Science, Middle East Technical University, 2021.