No-reference image quality assessment for high dynamic range images

Measuring digital picture quality, as perceived by human observers, is increasingly important in many applications in which humans are the ultimate consumers of visual information. Standard dynamic range (SDR) images provide 8 b/color/pixel. High dynamic range (HDR) images, usually created from multiple exposures of the same scene, can provide 16 or 32 b/color/pixel, but need to be tonemapped to SDR for display on standard monitors. Multiexposure fusion (MEF) techniques bypass HDR creation by fusing an exposure stack directly to SDR images to achieve aesthetically pleasing luminance and color distributions. Many HDR and MEF databases have a relatively small number of images and human opinion scores, obtained under stringently controlled conditions, thereby limiting realistic viewing. Moreover, many of these databases are intended to compare tone-mapping algorithms, rather than being specialized for developing and comparing image quality assessment models. To overcome these challenges, we conducted a massively crowdsourced online subjective study. The primary contributions described in this paper are: 1) the new ESPL-LIVE HDR Image Database that we created containing diverse images obtained by tone-mapping operators and MEF algorithms, with and without post-processing; 2) a large-scale subjective study that we conducted using a crowdsourced platform to gather more than 300 000 opinion scores on 1811 images from over 5000 unique observers; and 3) a detailed study of the correlation performance of the state-of-the-art no-reference image quality assessment algorithms against human opinion scores of these images. The database is available at http://signal.ece.utexas.edu/%7Edebarati/HDRDatabase.zip.

High Dynamic Range (HDR) reconstruction is the process of producing an HDR image from a set of Standard Dynamic Range (SDR) images with different exposure times. This is a particularly challenging problem when relative camera or object motion exists between the available SDR images. Recently, deep learning methods, specifically those based on convolutional neural networks (CNNs) have been developed for HDR and shown to achieve unprecedented quality gains. Invariably an image alignment phase precedes the CNN mapping and merging. In practice, this alignment step greatly increases the computational burden of deep HDR methods often rendering them unsuitable for real-time composition. We propose a new deep HDR technique that does not need any explicit alignment of SDR images. Instead, a novel attention mask is developed that enables the network to focus on parts of the scene with considerable motion. Further, a dense merger is proposed that leads to an economical network. Evaluation over benchmark databases reveals that the proposed AttenDense network achieves high quality HDR results with significantly reduced computation time than state of the art. Further, the incorporation of domain knowledge (development of a custom attention mask) allows a more graceful decay in performance in the face of limited training.

Due to the standard dynamic range (SDR) feature of those widely used displays, high dynamic range (HDR) images need to be tone-mapped for its visualization. Tone-mapping operators (TMOs) convert HDR image to different SDR images, thus the perceptual information loss of these tone-mapped images during the process of tone-mapping become a natural question. In this work, an objective method based on measuring the entropy loss between HDR and SDR images is proposed. We tested several different approach of this method, and determined which is better by calculating their correlation with subjective mean opinion score (MOS) conducted in a tone-mapped images databased (TMID). This method is then compared and combined with other full-reference tone-mapping image assessment methods: TMQI, FSITM and CDMI. Only linear combinations of these methods are tested since this method is a practical way, and the combination with best correlation was selected as the final quality score.

High Dynamic Range (HDR) imaging captures a wider range of luminance and color than Standard Dynamic Range (SDR), enabling more faithful representations of real-world scenes. Since most displays only support SDR, HDR content must be converted to SDR, often resulting in perceptual distortions like unnatural contrast, loss of detail or color artifacts. This paper evaluates No-Reference Image Quality Assessment (NR-IQA) methods on SDR images derived from HDR using the ESPL-LIVE HDR Image Database, which includes tone-mapped images and images created by multi-exposure fusion, together with associated mean opinion scores (MOS) and standard deviations, providing a reliable ground truth for evaluation. Seven NR-IQA methods, both traditional and Deep Learning (DL)-based, were tested and compared. DL algorithms performed best, with prediction errors below the standard deviation of MOS. Despite moderate correlation values (around 0.5), the results demonstrate the potential of these methods for practical image quality monitoring.

Printing display images normally proceeds well provided they comply with the SDR gamut supported by color management systems (CMS), like ICM used in Windows OS. Twilight vision, unfortunately, complicates this process, as it is quite difficult to convert images captured by UHDR systems given that their contrast ratios (CR) sometime exceeded 10^6:1 or so; SDR images have CR values of approx. 100:1. A typical example is the reproduction of lunar texture. Quite recently, we resolved this problem by employing just global tone mapping (GTM). Even with the use of GTM, emissive displays like LCD, not projection displays or printed matter, are strongly recommended to prevent image quality degradation. This study addresses the reasons why CMS fail to handle twilight vision material well and proposes enhanced GTM for printing emissive display images. The main point is the difference between the perception responses demonstrated in daytime and in twilight, even after the UHDR material is converted into SDR images. In other words, the key idea is to focus on the underutilized top and bottom margins of the reproduced SDR images. From another viewpoint, the most important result is the acquisition of super sensitivity. Our proposal enables high sensitivity even with the same sensor (increase is approx. 10 to 30 times).

In this chapter, depth estimation for stereo pair of High Dynamic Range (HDR) images is proposed. The proposed algorithm consists of two major techniques namely conversion of HDR images to Low Dynamic Range (LDR) images or Standard Dynamic Range (SDR) images and estimating the depth from the converted LDR / SDR stereo images. Local based tone mapping technique is used for the conversion of the HDR images to SDR images. And the depth estimation is done based on the corner features of the stereo pair images and block matching algorithm. Computationally much less expensive cost functions Mean Square Error (MSE) or Mean Absolute Difference (MAD) can be used for block matching algorithms. The proposed algorithm is explained with illustrations and results.

High Dynamic Range (HDR) imaging is gaining increased attention due to its realistic content, for not only regular displays but also smartphones. Before sufficient HDR content is distributed, HDR visualization still relies mostly on converting Standard Dynamic Range (SDR) content. SDR images are often quantized, or bit depth reduced, before SDR-to-HDR conversion, e.g. for video transmission. Quantization can easily lead to banding artefacts. In some computing and/or memory I/O limited environment, the traditional solution using spatial neighborhood information is not feasible. Our method includes noise generation (offline) and noise injection (online), and operates on pixels of the quantized image. We vary the magnitude and structure of the noise pattern adaptively based on the luma of the quantized pixel and the slope of the inverse-tone mapping function. Subjective user evaluations confirm the superior performance of our technique.

Abstract. 3D reconstruction relies on accurate detection, extraction, description and matching of image features. This is even truer for complex architectural scenes that pose needs for 3D models of high quality, without any loss of detail in geometry or color. Illumination conditions influence the radiometric quality of images, as standard sensors cannot depict properly a wide range of intensities in the same scene. Indeed, overexposed or underexposed pixels cause irreplaceable information loss and degrade digital representation. Images taken under extreme lighting environments may be thus prohibitive for feature detection/extraction and consequently for matching and 3D reconstruction. High Dynamic Range (HDR) images could be helpful for these operators because they broaden the limits of illumination range that Standard or Low Dynamic Range (SDR/LDR) images can capture and increase in this way the amount of details contained in the image. Experimental results of this study prove this assumption as they examine state of the art feature detectors applied both on standard dynamic range and HDR images.

Intelligent driver assistance systems are becoming increasingly popular in modern passenger vehicles. A crucial component of intelligent vehicles is the ability to detect vulnerable road users (VRUs) for an early and safe response. However, standard imaging sensors perform poorly in conditions of strong illumination contrast, such as approaching a tunnel or at night, due to their dynamic range limitations. In this paper, we focus on the use of high-dynamic-range (HDR) imaging sensors in vehicle perception systems and the subsequent need for tone mapping of the acquired data into a standard 8-bit representation. To our knowledge, no previous studies have evaluated the impact of tone mapping on object detection performance. We investigate the potential for optimizing HDR tone mapping to achieve a natural image appearance while facilitating object detection of state-of-the-art detectors designed for standard dynamic range (SDR) images. Our proposed approach relies on a lightweight convolutional neural network (CNN) that tone maps HDR video frames into a standard 8-bit representation. We introduce a novel training approach called detection-informed tone mapping (DI-TM) and evaluate its performance with respect to its effectiveness and robustness in various scene conditions, as well as its performance relative to an existing state-of-the-art tone mapping method. The results show that the proposed DI-TM method achieves the best results in terms of detection performance metrics in challenging dynamic range conditions, while both methods perform well in typical, non-challenging conditions. In challenging conditions, our method improves the detection F2 score by 13%. Compared to SDR images, the increase in F2 score is 49%.

Most Convolution Neural Network (CNN) based object detectors, to date, have been optimized for accuracy and/or detection performance on datasets typically comprised of well exposed 8-bits/pixel/channel Standard Dynamic Range (SDR) images. A major existing challenge in this area is to accurately detect objects under extreme/difficult lighting conditions as SDR image trained detectors fail to accurately detect objects under such challenging lighting conditions. In this paper, we address this issue for the first time by introducing High Dynamic Range (HDR) imaging to object detection. HDR imagery can capture and process ≈13 orders of magnitude of scene dynamic range similar to the human eye. HDR trained models are therefore able to extract more salient features from extreme lighting conditions leading to more accurate detections. However, introducing HDR also presents multiple new challenges such as the complete absence of resources and previous literature on such an approach. Here, we introduce a methodology to generate a large scale annotated HDR dataset from any existing SDR dataset and validate the quality of the generated dataset via a robust evaluation technique. We also discuss the challenges of training and validating HDR trained models using existing detectors. Finally, we provide a methodology to create an out of distribution (OOD) HDR dataset to test and compare the performance of HDR and SDR trained detectors under difficult lighting condition. Results suggest that using the proposed methodology, HDR trained models are able to achieve 10 – 12% more accuracy compared to SDR trained models on real-world OOD dataset consisting of high-contrast images under extreme lighting conditions.

No-reference image quality assessment in contourlet domain

The migration from High Definition (HD) TV to Ultra High Definition (UHD) is already underway. In addition to an increase of picture spatial resolution, UHD will bring more color and higher contrast by introducing Wide Color Gamut (WCG) and High Dynamic Range (HDR) video. As both Standard Dynamic Range (SDR) and HDR devices will coexist in the ecosystem, the transition from Standard Dynamic Range (SDR) to HDR will require distribution solutions supporting some level of backward compatibility. This paper presents a new HDR content distribution scheme, named SL-HDR1, using a single layer codec design and providing SDR compatibility. The solution is based on a pre-encoding HDR-to-SDR conversion, generating a backward compatible SDR video, with side dynamic metadata. The resulting SDR video is then compressed, distributed and decoded using standard-compliant decoders (e.g. HEVC Main 10 compliant). The decoded SDR video can be directly rendered on SDR displays without adaptation. Dynamic metadata of limited size are generated by the pre-processing and used to reconstruct the HDR signal from the decoded SDR video, using a post-processing that is the functional inverse of the pre-processing. Both HDR quality and artistic intent are preserved. Pre- and post-processing are applied independently per picture, do not involve any inter-pixel dependency, and are codec agnostic. Compression performance, and SDR quality are shown to be solidly improved compared to the non-backward and backward-compatible approaches, respectively using the Perceptual Quantization (PQ) and Hybrid Log Gamma (HLG) Opto-Electronic Transfer Functions (OETF).

We conducted a large-scale study of human perceptual quality judgments of High Dynamic Range (HDR) and Standard Dynamic Range (SDR) videos subjected to scaling and compression levels and viewed on three different display devices. While conventional expectations are that HDR quality is better than SDR quality, we have found subject preference of HDR versus SDR depends heavily on the display device, as well as on resolution scaling and bitrate. To study this question, we collected more than 23,000 quality ratings from 67 volunteers who watched 356 videos on OLED, QLED, and LCD televisions, and among many other findings, observed that HDR videos were often rated as lower quality than SDR videos at lower bitrates, particularly when viewed on LCD and QLED displays. Since it is of interest to be able to measure the quality of videos under these scenarios, e.g. to inform decisions regarding scaling, compression, and SDR vs HDR, we tested several well-known full-reference and no-reference video quality models on the new database. Towards advancing progress on this problem, we also developed a novel no-reference model called HDRPatchMAX, that uses a contrast-based analysis of classical and bit-depth features to predict quality more accurately than existing metrics.

Tone mapping operators (TMO) are functions that map high dynamic range (HDR) images to a standard dynamic range (SDR), while aiming to preserve the perceptual cues of a scene that govern its visual quality. Despite the increasing number of studies on quality assessment of tone mapped images, current subjective quality datasets have relatively small numbers of images and subjective opinions. Moreover, existing challenges in transferring laboratory experiments to crowdsourcing platforms put a barrier for collecting large-scale datasets through crowdsourcing. In this work, we address these challenges and propose the RealVision-TMO (RV-TMO), a large-scale tone mapped image quality dataset. RV-TMO contains 250 unique HDR images, their tone mapped versions obtained using four TMOs and pairwise comparison results from seventy unique observers for each pair. To the best of our knowledge, this is the largest dataset available in the literature for quality evaluation of TMOs by the number of tone mapped images and number of annotations. Furthermore, we provide a content selection strategy to identify interesting and challenging HDR images. We also propose a novel methodology for observer screening in pairwise experiments. Our work does not only provide annotated data to benchmark existing objective quality metrics, but also paves the path to building new metrics for tone mapping quality evaluation.

Evaluating the quality of high dynamic range (HDR) images has emerged as a challenging and contemporary topic with the proliferation of HDR content. In this work, a new HDR compression (HDRC) database is proposed, aiming to provide a benchmark for the development of full-reference HDR image quality assessment (IQA) algorithms when facing the latest HDR compression distortions. In particular, the proposed HDRC database is the first HDR-IQA database to incorporate Versatile Video Coding (VVC) compression distortions, closely associated with practical application scenarios. Furthermore, the proposed HDRC database is currently the largest HDR-IQA database, including 80 reference images and 400 distorted images. Extensive experiments are conducted by studying the performance compared to existing HDR-IQA databases when evaluating three HDR-specific IQA models and nine IQA models prevalent for low dynamic range (LDR) content, revealing the challenges the proposed HDRC database brings. The results indicate that the existing IQA models demonstrate noticeable decreases in accuracy when assessing new compression distortions, underscoring the need for the development of novel HDR-IQA models. Consequently, the suggested HDRC database can serve as a potential database for HDR-IQA research, fostering a comprehensive exploration of the associated fields.