Deinterlacing Using Variational Methods

Abstract

We present a variational framework for deinterlacing that was originally used for inpainting and subsequently redeveloped for deinterlacing. From the framework, we derive a motion adaptive (MA) deinterlacer and a motion compensated (MC) deinterlacer and test them together with a selection of known deinterlacers. To illustrate the need for MC deinterlacing, the problem of details in motion (DIM) is introduced. It cannot be solved by MA deinterlacers or any simpler deinterlacers but only by MC deinterlacers. The major problem in MC deinterlacing is computing reliable optical flow [motion estimation (ME)] in interlaced video. We discuss a number of strategies for computing optical flows on interlaced video hoping to shed some light on this problem. We produce results on challenging real world video data with our variational MC deinterlacer that in most cases are indistinguishable from the ground truth.

Partial Differential Equations for Zooming, Deinterlacing and Dejittering

“…by using standard interpolation schemes such as piecewise constant (line doubling), linear and cubic interpolation. State-of-the-art Ballester et al 2007;Ghodstinat et al 2009;Keller et al 2005Keller et al , 2008Tschumperle and Besserer 2004).…”

Efficient deinterlacing method using simple edge slope tracing

“…[1][2][3][4][5][6][7][8][9][10][11][12][13][14][15] In general, TV broadcasters choose interlaced scanning as the main standard for video formats. 16 Since it alternately scans two fields by dividing a frame into two parts, interlacing may cause video quality degradation such as flickering.…”

Deinterlacing with Motion-Compensated Anisotropic Diffusion

“…In contrast to their motion adaptive counterparts these techniques make use of a real motion estimator. Thus they are able to correct the image sequence by the occurring motion either before or during the actual interpolation step [1,4,10,23,29].…”

“…Due to their ability to consider motion information in the deinterlacing process, motion adaptive and motion compensated approaches give the best results out of these four classes [2,22,23,27]. However, both strategies can hardly be compared, since they have quite complementary advantages and shortcomings: While motion adaptive techniques can provide very accurate results, they are known to have severe problems with large displacements [22].…”