Image quality evaluation in the field of digital film restoration

Chambah, Majed; Saint-Jean, Christophe; Helt, Francois

doi:10.1117/12.586738

Cited by 3 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unlike tasks such as JPEG artefact removal, super‐resolution or colourisation, there are no robust simulation models of analogue film damage to generate high‐quality synthetic training samples. There is also no consensus as to what makes a “good” restoration, and therefore how a film restoration model should be evaluated, especially without human‐restored ground‐truth scans [CSJH05, Cha19].…”

Section: Introductionmentioning

confidence: 99%

Simulating analogue film damage to analyse and improve artefact restoration on high‐resolution scans

Ivanova

Williamson

Henderson

2023

Computer Graphics Forum

View full text Add to dashboard Cite

Digital scans of analogue photographic film typically contain artefacts such as dust and scratches. Automated removal of these is an important part of preservation and dissemination of photographs of historical and cultural importance. While state‐of‐the‐art deep learning models have shown impressive results in general image inpainting and denoising, film artefact removal is an understudied problem. It has particularly challenging requirements, due to the complex nature of analogue damage, the high resolution of film scans, and potential ambiguities in the restoration. There are no publicly available high‐quality datasets of real‐world analogue film damage for training and evaluation, making quantitative studies impossible. We address the lack of ground‐truth data for evaluation by collecting a dataset of 4K damaged analogue film scans paired with manually‐restored versions produced by a human expert, allowing quantitative evaluation of restoration performance. We have made the dataset available at https://doi.org/10.6084/m9.figshare.21803304. We construct a larger synthetic dataset of damaged images with paired clean versions using a statistical model of artefact shape and occurrence learnt from real, heavily‐damaged images. We carefully validate the realism of the simulated damage via a human perceptual study, showing that even expert users find our synthetic damage indistinguishable from real. In addition, we demonstrate that training with our synthetically damaged dataset leads to improved artefact segmentation performance when compared to previously proposed synthetic analogue damage overlays. The synthetically damaged dataset can be found at https://doi.org/10.6084/m9.figshare.21815844, and the annotated authentic artefacts along with the resulting statistical damage model at https://github.com/daniela997/FilmDamageSimulator. Finally, we use these datasets to train and analyse the performance of eight state‐of‐the‐art image restoration methods on high‐resolution scans. We compare both methods which directly perform the restoration task on scans with artefacts, and methods which require a damage mask to be provided for the inpainting of artefacts. We modify the methods to process the inputs in a patch‐wise fashion to operate on original high resolution film scans.

show abstract

Section: Introductionmentioning

confidence: 99%

Simulating analogue film damage to analyse and improve artefact restoration on high‐resolution scans

Ivanova

Williamson

Henderson

2023

Computer Graphics Forum

View full text Add to dashboard Cite

show abstract

“…This explains why in some domains such as photography and old film restoration, where there is no reference to compare to, subjective quality evaluation is the most reliable technique used. In the last works we presented some original reference free metrics [22][23][24][25] that correlated well with human perception (metrics for contrast and color quality based on statistical and perceptual approaches). These perception correlated metrics were a first step to bridge the gap between subjective and automatic approaches.…”

mentioning

confidence: 98%

Toward an automatic subjective image quality assessment system

et al. 2009

Self Cite

View full text Add to dashboard Cite

Usually in the field of image quality assessment the terms "automatic" and "subjective" are often incompatible. In fact, when it comes to image quality assessment, we have mostly two kinds of evaluation techniques: subjective evaluation and objective evaluation. Only objective evaluation techniques being automatizable, while subjective evaluation techniques are performed by a series of visual assessment done by expert or non-expert observers. In this paper, we will present a first attempt to an automatic subjective quality assessment system. The system computes some perception correlated color metrics from a learning set of images. During the learning stage a subjective assessment by users is required so that the system matches the subjective opinions with computed metrics on a variety of images. Once the learning process is over, the system operates in an automatic mode using only the learned knowledge and the reference free computed metrics from the images to assess. Results and also future prospects of this work are presented. CONTEXTWhen it comes to image quality assessment, we have mostly two kinds of evaluation techniques [20] : subjective evaluation and objective evaluation. These evaluations can be done with a priori information (with reference) or without use of a priori information (reference free). Subjective approaches, which to date are the only widely recognized method of determining actual perceived quality, are complex and time-consuming, both in their preparation and execution [21]. Subjective evaluation is formalized with defined procedures [20]. Objective quality evaluation use metrics to evaluate image quality. Objective evaluation is automated, hence it costs less than a subjective evaluation, plus it can be done in real-time since it needs no user interaction. Objective quality metrics can be full reference, reduced reference or reference free [21] Moreover, automatic objective assessment systems do not necessarily correlate well with perceived quality [20,21]. Ideally, a quality assessment system would perceive and measure image or video impairments just like a human being. As long as quality metrics are not correlated with human perception, subjective evaluation is still mandatory. This explains why in some domains such as photography and old film restoration, where there is no reference to compare to, subjective quality evaluation is the most reliable technique used. In the last works we presented some original reference free metrics [22][23][24][25] that correlated well with human perception (metrics for contrast and color quality based on statistical and perceptual approaches). These perception correlated metrics were a first step to bridge the gap between subjective and automatic approaches. In fact, since perceptual correlated metrics are available, automatic computation and learning -and hence automatic assessment-become possible.In this paper, we will present a first attempt to an automatic subjective quality assessment system. The system computes some perception correlated color me...

show abstract

Controlled and uncontrolled viewing conditions in the evaluation of prints

Brambilla¹,

Eschbach²,

Rizzi³

2008

SPIE Proceedings

View full text Add to dashboard Cite

Visual experiments, attesting visual preference, visual ranking and visual differentiation, are very important to academia and industry. They are traditionally performed into laboratories under controlled viewing conditions, resulting very costly in their execution, due to the time and effort involved by all participants. If controlled tests could be substituted by uncontrolled tests, a potential serious improvement could be obtained by eliminating a large part of the cost. In this work we investigate if, and to what extent, visual experiments performed under controlled viewing conditions can be substituted by uncontrolled experiments. A task of visual preference of prints is carried out. This task is performed in the laboratory, under controlled viewing conditions, and in many different places, under natural, artificial and mixed light. We observe statistical equivalence for preferences expressed in controlled or uncontrolled conditions that supports the hypothesis that visual preference can be assessed with uncontrolled tests.

show abstract

Image quality evaluation in the field of digital film restoration

Cited by 3 publications

References 0 publications

Simulating analogue film damage to analyse and improve artefact restoration on high‐resolution scans

Simulating analogue film damage to analyse and improve artefact restoration on high‐resolution scans

Toward an automatic subjective image quality assessment system

Controlled and uncontrolled viewing conditions in the evaluation of prints

Contact Info

Product

Resources

About