An HVS-inspired video deinterlacer based on visual saliency

Abstract

Video deinterlacing is a technique wherein the interlaced video format is converted into progressive scan format for nowadays display devices. In this paper, a spatial saliency-guided motion-compensated deinterlacing method is proposed which accounts for the properties of the Human Visual System (HVS): our algorithm classifies the field according to its texture and viewer's region of interest and adapts the motion estimation and compensation, as well as the saliency-guided interpolation to ensure high-quality frame reconstruction. Two different saliency models, namely the graph-based visual saliency (GBVS) model and the spectral residual visual saliency (SRVS) model, have been studied and compared in terms of visual quality performances as well as computational complexity. The experimental results on a great variety of video test sequences show significant improvement of reconstructed video quality with the GBVS-based proposed method compared to classical motion-compensated and adaptive deinterlacing techniques, with up to 4.5 dB gains in terms of PSNR. Simulations also show that the SRVS-based deinterlacing process can result to significant reductions of complexity (up to 25 times a decrease of the computation time compared with the GBVSbased method) at the expense of a PSNR decrease.

Image Abstraction Framework as a Pre-processing Technique for Accurate Classification of Archaeological Monuments Using Machine Learning Approaches

“…In some circumstances, it is necessary to integrate the reverse tone mapping operator between the hybrid TMO in order to get the best esthetic tone mapping result. Against this background, this work adopted the K-TMO which is a hybrid of local and global tonality mapping operators consisting of Ashikhmin local TMO [80], Gamma TMO [81] and Banterle global TMO [82] jointly with and deinterlacing method [83]. Ashikhmin TMO is superior in regulating and retaining the local-luminance in the whole image.…”

Rethinking deinterlacing for early interlaced videos

“…Because interlaced aliasing is more significant than blur of interpolation, various single field interpolation methods have been proposed in the past decades, for instance, edge detection with fuzzy logic [3], block-wise autoregression [4], bilateral filtering model [5], locality and similarity adaption [6], the maximum a posteriori (MAP) model [7] and least square methods [8]. To further refine the visual quality, visual saliency is also introduced to improve deinterlacing, for example, spectral residual visual saliency-based deinterlacing [9], [10]. There are also some single field interpolation methods for special applications, such as GPU-parallel implementation of edge-directed interpolation [11] and deinterlacing for 1080p-to-8KUHD [12].…”