Tone mapping

About: Tone mapping is a research topic. Over the lifetime, 1713 publications have been published within this topic receiving 48490 citations.

Tone Mapping Techniques and Color Image Difference in Global Illumination

Abstract: Tone mapping is the final step of every rendering process. Due to display devices’ nonlinearities, reduced color gamuts and moderate dynamic ranges it is necessary to apply some mapping technique on the computed radiances. We described mapping methods that are considered to be state of the art today, and some newly developed techniques. The main contributions of this thesis in tone mapping techniques are interactive calibration of contrast and aperture, minimum information loss methods and incident light metering. The interactive calibration technique makes it possible to display a desired scene lighting atmosphere if the radiance values are rendered in fictitious units. Minimum information loss techniques are based, in a way, on the photographers’ approach. The mapping function is applied only on a certain radiance interval, which is chosen automatically. The original contrast of all pixels inside the interval is preserved. Furthermore, the bounded error version of the minimum loss method is an extension of Schlick’s method. The incident light metering method was inspired by the photographers’ approach, too. This method makes it possible to reproduce original colors faithfully. Even if the average reflectance of a scene is very low, or very high, this method will reproduce original colors, which is not the case with other methods. The idea is to measure the incident light using diffusors in the scene, and then to compute a scale factor based on the incident light and apply this scale factor on the computed radiances. Beside these, other tone mapping techniques are described in this work. We described Tumblin and Rushmeier’s mapping, Ward’s contrast based scale factor, the widely used mean value mapping, an exponential mapping introduced by Ferschin et al., Schlick’s mapping, a visibility matching tone operator introduced by Larson et al., and a model of visual adaptation proposed by Ferwerda et al. Unfortunately there is no ultimative solution to the tone mapping problem. Every method has its strengths and weaknesses, and the user should choose a method according to his or her needs. Finally, this thesis ends with a color image difference algorithm. A good image metric is often needed in computer graphics. The method described here is a perception based metric that operates in the original image space (there is no need for Fourier or wavelet transform), what makes the whole method fast and intuitive. This is the only method that stresses distance dependency explicitly.

Tone mapping using color correction function and image decomposition in high dynamic range imaging

Abstract: In this paper, we propose a tone mapping (TM) method using color correction function (CCF) and image decomposition in high dynamic range (HDR) imaging. The CCF in the proposed TM is derived from the luminance compression function with the color constraint under which the color ratios, between the three color channels of the radiance map and dynamic range compression term, are preserved and color saturation is controlled. The proposed CCF is developed to locally perform the luminance compression and color saturation control in local TM. For image decomposition, we use a bilateral filter and apply the adaptive weight to the base layer of the luminance. Computer simulations with various sets of real, low dynamic range images show the effectiveness of the proposed TM algorithm in terms of the visual quality as well the local contrast. It can be used for contrast and color enhancement in various display and acquisition devices.

A model of local adaptation

Abstract: The visual system constantly adapts to different luminance levels when viewing natural scenes. The state of visual adaptation is the key parameter in many visual models. While the time-course of such adaptation is well understood, there is little known about the spatial pooling that drives the adaptation signal. In this work we propose a new empirical model of local adaptation, that predicts how the adaptation signal is integrated in the retina. The model is based on psychophysical measurements on a high dynamic range (HDR) display. We employ a novel approach to model discovery, in which the experimental stimuli are optimized to find the most predictive model. The model can be used to predict the steady state of adaptation, but also conservative estimates of the visibility (detection) thresholds in complex images. We demonstrate the utility of the model in several applications, such as perceptual error bounds for physically based rendering, determining the backlight resolution for HDR displays, measuring the maximum visible dynamic range in natural scenes, simulation of afterimages, and gaze-dependent tone mapping.

TMO-Net: A Parameter-Free Tone Mapping Operator Using Generative Adversarial Network, and Performance Benchmarking on Large Scale HDR Dataset

Abstract: Currently published tone mapping operators (TMO) are often evaluated on a very limited test set of high dynamic range (HDR) images. Thus, the resulting performance index is highly subject to extensive hyperparameter tuning, and many TMOs exhibit sub-optimal performance when tested on a broader spectrum of HDR images. This indicates that there are deficiencies in the generalizable applicability of these techniques. Finally, it is a challenge developing parameter-free tone mapping operators using data-hungry advanced deep learning methods due to the paucity of large scale HDR datasets. In this paper, these issues are addressed through the following contributions: a) a large scale HDR image benchmark dataset (LVZ-HDR dataset) with multiple variations in sceneries and lighting conditions is created to enable performance evaluation of TMOs across a diverse conditions and scenes that will also contribute to facilitate the development of more robust TMOs using state-of-the-art deep learning methods; b) a deep learning-based tone mapping operator (TMO-Net) is presented, which offers an efficient and parameter-free method capable of generalizing effectively across a wider spectrum of HDR content; c) finally, a comparative analysis, and performance benchmarking of 19 state-of-the-art TMOs on the new LVZ-HDR dataset are presented. Standard metrics including the Tone Mapping Quality Index (TMQI), Feature Similarity Index for Tone Mapped images (FSITM), and Natural Image Quality Evaluator (NIQE) are used to qualitatively evaluate the performance index of the benchmarked TMOs. Experimental results demonstrate that the proposed TMO-Net qualitatively and quantitatively outperforms current state-of-the-art TMOs.

Image Demoireing with Learnable Bandpass Filters

Abstract: Image demoireing is a multi-faceted image restoration task involving both texture and color restoration. In this paper, we propose a novel multiscale bandpass convolutional neural network (MBCNN) to address this problem. As an end-to-end solution, MBCNN respectively solves the two sub-problems. For texture restoration, we propose a learnable bandpass filter (LBF) to learn the frequency prior for moire texture removal. For color restoration, we propose a two-step tone mapping strategy, which first applies a global tone mapping to correct for a global color shift, then performs local fine tuning of the color per pixel. Through an ablation study, we demonstrate the effectiveness of the different components of MBCNN. Experimental results on two public datasets show that our method outperforms state-of-the-art methods by a large margin (more than 2dB in terms of PSNR).