Demosaicing remains a critical component of modern digital image processing, with a direct impact on image quality. Conventional demosaicing methods yield relatively poor results especially light-weight methods used for fast processing. Alternatively, recent works utilizing Deep Convolutional Neural Nets have significantly improved upon previous methods, increasing both quantitative and perceptual performance. This approach has seen significant reduction of artifacts but there still remains scope for meaningful improvement. To further this research, we investigate the use of alternate architectures and training parameters to reduce incurred errors, especially visually disturbing demosaicing artifacts such as moiré and provide an overview of current methods to better understand their expected performance. Our results show a U-NET style Network to outperform previous methods in quantitative and perceptual error and remain computationally efficient for use in GPU accelerated applications as an end-to-end demosaicing solution.
Motion estimation is a key component of any modern video codec. Our understanding of motion and the estimation of motion from video has come a very long way since 2000. More than 135 different algorithms have been recently reviewed by Scharstein et al http://vision.middlebury.edu/flow/. These new algorithms differ markedly from Block Matching which has been the mainstay of video compression for some time. This paper presents comparisons of H.264 and MP4 compression using different motion estimation methods. In so doing we present as well methods for adapting pre-computed motion fields for use within a codec. We do not observe significant gains to be had with the methods chosen w.r.t. Rate Distortion tradeoffs but the results reflect a significantly more complex interrelationship between motion and compression than would be expected. There remains much more to be done to improve the coverage of this comparison to the emerging standards but these initial results show that there is value in these explorations.
Temporal and spatial random variation of luminance in images, or
'flicker' is a typical degradation observed in archived film and
video. The underlying premise in typical flicker reduction algorithms is that each image must be corrected for a spatially varying gain and
offset. These parameters are estimated in the stationary region of the
image. Hence the performance of that algorithm depends crucially on the identification of stationary image regions. Position fluctuations are also a common artefact resulting in a random 'shake' of each film frame. For removing both, the key is to reject regions showing local motion or other outlier activity. Parameters are then estimated mostly on that part of the image undergoing the dominant motion. A new
algorithm that simultaneously deals with global motion estimation and
flicker is presented. The final process is based on a robust application of weighted least-squares, in which the weights also classify portions of the image as local or global. The paper presents results on severely degraded sequences showing evidence of both Flicker and random shake.