Perception of noise and global illumination: Toward an automatic stopping criterion based on SVM

Takouachet, Nawel; Delepoulle, Samuel; Renaud, Christophe; Zoghlami, Nesrine; Tavares, João Manuel R. S.

doi:10.1016/j.cag.2017.09.008

Cited by 9 publications

(18 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The difference between the original image and the noise-free image, so-called “noise mask” is used directly as the input to the model, i.e., a vector of size

. Another approach proposed in [ 42 ], consists in adding samples to the input image using path tracing algorithm in order to obtain an image considered as a reference. For each of the two images (input image and approximate reference) a blurred image is computed using a

Gaussian convolution with a convolution coefficient

.…”

Section: Previous Workmentioning

confidence: 99%

See 1 more Smart Citation

Stopping Criterion during Rendering of Computer-Generated Images Based on SVD-Entropy

Buisine

Bigand

Synave

et al. 2021

Entropy

Self Cite

View full text Add to dashboard Cite

The estimation of image quality and noise perception still remains an important issue in various image processing applications. It has also become a hot topic in the field of photo-realistic computer graphics where noise is inherent in the calculation process. Unlike natural-scene images, however, a reference image is not available for computer-generated images. Thus, classic methods to assess noise quantity and stopping criterion during the rendering process are not usable. This is particularly important in the case of global illumination methods based on stochastic techniques: They provide photo-realistic images which are, however, corrupted by stochastic noise. This noise can be reduced by increasing the number of paths, as proved by Monte Carlo theory, but the problem of finding the right number of paths that are required in order to ensure that human observers cannot perceive any noise is still open. Until now, the features taking part in the human evaluation of image quality and the remaining perceived noise are not precisely known. Synthetic image generation tends to be very expensive and the produced datasets are high-dimensional datasets. In that case, finding a stopping criterion using a learning framework is a challenging task. In this paper, a new method for characterizing computational noise for computer generated images is presented. The noise is represented by the entropy of the singular value decomposition of each block composing an image. These Singular Value Decomposition (SVD)-entropy values are then used as input to a recurrent neural network architecture model in order to extract image noise and in predicting a visual convergence threshold of different parts of any image. Thus a new no-reference image quality assessment is proposed using the relation between SVD-Entropy and perceptual quality, based on a sequence of distorted images. Experiments show that the proposed method, compared with experimental psycho-visual scores, demonstrates a good consistency between these scores and stopping criterion measures that we obtain.

show abstract

“…The difference between the original image and the noise-free image, so-called “noise mask” is used directly as the input to the model, i.e., a vector of size

Gaussian convolution with a convolution coefficient

.…”

Section: Previous Workmentioning

confidence: 99%

“…It is thus very different of well-known noise models and using available natural image databases is not possible. Then data that were used in [ 41 , 42 ] are still limited both in terms of image size and scene complexity.…”

Section: Subjective Dataset Collectionmentioning

confidence: 99%

Stopping Criterion during Rendering of Computer-Generated Images Based on SVD-Entropy

Buisine

Bigand

Synave

et al. 2021

Entropy

Self Cite

View full text Add to dashboard Cite

show abstract

“…В работах [53] была представлена модель, анализирующая шум в изоб-ражениях, полученных методами Монте-Карло на основе визуального воспри-ятия человека при помощи машинного обучения. Данная модель использова-лась для выработки критерия остановки расчёта.…”

Section: 3unclassified

“…Поэтому задача точного отделения сигнала от шу-ма в Монте-Карло-рендеринге в общем случае -трудная. Недостатком работ [53,54] является то, что анализировались сцены только с ламбертовскими ма-териалами и простым освещением. Это сильно отличается от сцен с трудновы-числимыми феноменами освещённости, поэтому в данной работе аналогичный подход не был использован.…”

Section: 3unclassified

“…Нетрудно заметить, что используемый детектор шума основан на принци-пе работы известного медианного фильтра. Хотя в настоящий момент доступ-ны более современные методы детекции шума (например, [50][51][52][53][54]), было обна-ружено, что для синтезируемых при помощи небольшого числа OMC сэмплов изображений широкого динамического диапазона описанный метод детекции даёт приемлемую числовую оценку шума в интервале от 0 до 1. Более точная оценка не требуется (а зачастую она и невозможна, поскольку, как обсуждалось в обзоре предыдущих работ, шум есть результат неизвестного в общем случае распределения), т. к. далее происходит агрегация шума по области.…”

Section: построение функции значимостиunclassified

See 1 more Smart Citation

Selective application of Metropolis Light Transport for hard sampling illumination phenomena

Frolov

2017

KIAM Prepr.

View full text Add to dashboard Cite

Guided-Generative Network: A New Robust Deep Learning Architecture for Noise Characterization in Monte-Carlo Rendering

Buisine

Teytaud

Delepoulle

et al. 2022

Deep Learning Applications, Volume 4

Self Cite

View full text Add to dashboard Cite

Current methods for generating realistic computer-generated images rely on stochastic lighting simulation techniques based on a Monte Carlo approach. These Monte Carlo simulations construct light paths between the camera and light sources within the virtual 3D model to calculate the appearance of objects and provide realistic images. Insufficient sampling of the light path space results in high variance between individual pixels in the image which is visually perceived as noise. To reduce this noise, the number of light path samples must be increased until the noise is no longer visible, but this has the disadvantage of significantly increasing the computation time. Finding the right number of samples needed for human observers to perceive no noise remains an open problem that this paper addresses using a new neural network architecture called Guided-Generative Network (GGN). As it is often difficult to extract features from an image for classification tasks, the GGN attempts to automatically find the desired features for noise detection. This is done through an architecture composed of 3 models that collaborate to characterize the noise present and guide the classification. The results obtained show that the GGN can correctly solve the problem without prior knowledge of the noise while being competitive with existing methods. A visual validation experiment of the images obtained by the model indicates a significant reliability to the requested task.

show abstract

Perception of noise and global illumination: Toward an automatic stopping criterion based on SVM

Cited by 9 publications

References 40 publications

Stopping Criterion during Rendering of Computer-Generated Images Based on SVD-Entropy

Stopping Criterion during Rendering of Computer-Generated Images Based on SVD-Entropy

Selective application of Metropolis Light Transport for hard sampling illumination phenomena

Guided-Generative Network: A New Robust Deep Learning Architecture for Noise Characterization in Monte-Carlo Rendering

Contact Info

Product

Resources

About