Automatic Restoration of Old Motion Picture Films Using Spatiotemporal Exemplar-Based Inpainting

Abstract

Abstract. This paper presents a method for automatic removal of local defects such as blotches and impulse noise in old motion picture films. The method is fully automatic and includes the following steps: fuzzy prefiltering, motioncompensated blotch detection, and spatiotemporal inpainting. The fuzzy prefilter removes small defective areas such as impulse noise. Modified bidirectional motion estimation with a predictive diamond search is utilized to estimate the motion vectors. The blotches are detected by the rank-ordered-difference method. Detected missing regions are interpolated by a new exemplar-based inpainting approach that operates on three successive frames. The performance of the proposed method is demonstrated on an artificially corrupted image sequence and on a real motion picture film. The results of the experiments show that the proposed method efficiently removes flashing and still blotches and impulse noise from image sequences.

Repairing of Motion Vectors within Defect Regions Based on Temporal Continuity of Motion Trajectory and Blotch Detection in Old Color Films

“…Fakat bir onceki ,er,eve zaten onarildigi i,in, 3 numarali ok ile gosterilen bir onceki ,er,eveden bir sonraki ,er,eveye yapilan hareket kestirimi dogruluklu sonucu vereceginden yulrulrlulkteki blogun, 2 numarali ok ile gosterilen ileri zamansal yondeki hareket vektorui a*agidaki gibi hesaplanabilir: u( yt)= Ua(X,Y,t-1) (6) u(x,y,6) Ayni i,lem, 1 numarali ok ile gosterilen geri zamansal yon i,in de yapilir. WBylece lekeli bolgelerdeki imge bloklarinin hareket vektorleri, ,er,eveler arasi zamansal ilintiye bagli olarak bulunmu, olur.…”

“…ve ger,ek bozukluk i,eren imge dizisinin 58. film ,er,evelerinde kestirilen hareket vektorleri ve AGAI yontemiyle algilanmi* lekeler ve algilama ba*arimi, sirasiyla $ekil 4., $ekil 5. ve $ekil6. 'da gosterilmi,tir.…”

Archive Film Restoration Based on Spatiotemporal Random Walks

“…texture features such as the Local Binary Pattern (LBP) [20] (leading to consistent region representation). Unlike previous methods such as [19,23,16,10], we consider multiple features in an integrated fashion and show that this provides better restoration than treating the features separately. The computational expense incurred due to the use of more features is an affordable tariff in our archive restoration application where accuracy is of paramount importance.…”

A Spatiotemporal Algorithm for Detection and Restoration of Defects in Old Color Films

“…We demonstrate the algorithm performance on quantitative and qualitative manners and compare to the exemplar-based inpainting [4], the ML3Dex vector median [5] and the spatiotemporal search [6]. 0.1% Gauss noise and random size and shape blotches were artificially generated for the synthetic sequences, and frames 233 and 3 of "Foreman" and "Coastguard" are shown in Figs.…”

“…This approach successfully conceals blotches on image sequence if true motion estimation can be done. Moreover, Gangal et al [6] presented a spatiotemporal reconstruction algorithm which occasionally fails to fill in missing areas, involved the complex texture and structure.…”