belmar bevley daysia
Juan Navarro
In photography, Dynamic Range (DR) has different definitions depending on what we are talking about. The dynamic range of the scene is the ratio of the highest luminance to the lowest luminance in the field of view of the scene. Dynamic range is usually expressed in stops of light and can vary drastically!
As we can see, while we can capture high dynamic range content, the challenge is to fit that expansive range into the constraints of the medium. It requires condensing the vast range of light levels into the narrower range while keeping the perception of contrast. This is called HDR tone mapping.
Smartphone cameras capture HDR images using techniques like exposure bracketing, in which the same scene is captured in different exposures, and multiple frames merging, in which several captured frames are combined to form one image. Many image algorithms are involved and implemented thanks to the help of hardware accelerators called ISP (Image Signal Processing). During this process, images are stored on a very large number of bits (up to 18 bits on the latest ISP!).
With standard dynamic range (SDR) storage format and displays, the vast amount of information collected will be compressed into 8-bit code values, which can not fully represent the complete dynamic range. Consequently, this compression may result in a loss of contrast and detail in the final image. When viewed on an SDR display, only a portion of the scene can be faithfully reproduced, further limiting the ability to showcase the full richness of the moment.
With the introduction of different technologies, such as OLED, or local dimming for LCDs, the screen dynamic range has increased. The brightest pixel can go very bright, while the darkest one (if no reflection occurs on your screen) can go very low. The measure of display luminance is expressed in candelas per square meter (cd/m), which is sometimes referred to by its non-SI name, nits. In the past, a typical display could achieve a maximum luminance of about 200 cd/m (nits). Today, some advanced monitors can support 1000 cd/m (nits) across their entire screen, with peak luminance reaching up to 2000 cd/m (nits) This is a significant increase in brightness compared to traditional displays.
The 10-bit input is one of the reasons why to fully benefit from the performance of the display, one needs to define new image formats, known as HDR Photo formats. These file formats contains 10 bits of data, but also some side data (aka metadata) to help the playback system interpret correctly the content to be displayed, knowing the characteristics of the screen.
HDR video can be delivered using different Electro-Optical Transfer Functions (EOTFs), color primaries and metadata type. These different EOTFs, color primaries or metadata are standardized and published as recommendations by organizations such as SMPTE and ITU. For example, the Perceptual Quantizer (PQ) EOTF is standardized by SMPTE in ST-2084 as well as by ITU in Rec. 2100, HLG as a transfer function is standardized by ITU in Rec. 2100. The color primaries and viewing conditions are also standardized by the ITU in Rec. 2100 and Rec. 2020.
Using these recommendations, there are different HDR formats in which video contents are encoded and delivered. HDR10+, Dolby Vision, HDR Vivid, etc. are all different HDR formats that use dynamic metadata (SMPTE ST 2094) where the entire video uses the metadata on a scene-by-scene basis, trying to preserve the artistic intent to the greatest extent. HDR10 uses static metadata (SMPTE 2086), thus the same metadata is used for each frame of the video. HLG (based on the HLG EOTF) is another HDR format that has no metadata requirements but is backward compatible with SDR content.
On a side note, the viewing environment is also a limiting factor to fully enjoy the HDR experience. In particular, the environment is driving the human eye adaptation, therefore changing our perception of dark and bright levels. This is the reason why grading studios use strictly standardized lighting conditions when using HDR monitors. Smartphones also propose HDR displays, but the viewing environment is much less controlled. The adaptation of the smartphone display remains to this day one of the main challenges for HDR, as we shall see in a future article.
In the world of visual displays, a reference white is like a standard benchmark for brightness, and it serves as the foundation for all other colors. The white background of this web page (provided you are in non-dark mode) is the reference white of the screen.
Regardless of the specific HDR standard in use, a fundamental goal is to extend the displayed dynamic range. Conventional displays have faced challenges when it comes to rendering details, particularly in highlights. On a standard display, the reference white adheres to specifications at 100 cd/m. It might be logical to assume that on a good HDR display, this reference white would shine at well over 1000 cd/m, given the focus on brightness.
This expanded range allows tones and colors to have more space to express themselves, resulting in brighter highlights, deeper shadows, enhanced tonal distinctions, and more vibrant colors. The outcome is that photos optimized for HDR displays deliver a heightened impact and a heightened sense of depth and realism, making the visual experience far more immersive and captivating. However, the appearance of HDR content is susceptible to variations across different devices, owing to the diverse capabilities of HDR displays and the distinct tone mapping methods employed by various software and platforms.
Recalling the luminance scale, typical SDR images, define black and white as 0.2 and 100 cd/m, respectively. In contrast, HDR images define black and a default reference white as 0.0005 and 203 cd/m, respectively, signifying that everything above 203 cd/m is considered headroom.
In 2020, Apple pioneered the use of gain maps with the HEIC image format. The iPhone images have incorporated additional data that enables the reconstruction of an HDR representation from the original SDR image. This approach has now been standardized in the iOS 17 release [2].
Google, with the Android 14 release also implements a gain map method called Ultra HDR [3], while the gain map specification published by Adobe [4], provides a formal description of the gain map approach for storing, displaying, and tone mapping HDR images.
As a result, the image displayed on an HDR monitor will look more realistic and detailed than the image displayed on a traditional monitor. Here is an illustrative animation, designed to be seen on SDR displays, of the image enhancement when transitioning from SDR to HDR display. Beware, this is only a simulation! The transition between an SDR image and an HDR image on an HDR display with a proper setup would be more impressive!
As another example, laptops have morphed into personal entertainment centers, allowing us to enjoy listening to music and watching movies. But here, too, discrepancies arise. Some laptops offer high-quality audio, while others leave us wanting more. Display color accuracy and brightness also vary, impacting enjoyment.
As with our other protocols, our testing philosophy for laptop evaluation is centered on how people use their laptops and the features that are most important to them. We research this information through our own formal surveys and focus groups, in-depth studies of customer preferences conducted by manufacturers, and interviews with imaging and sound professionals.
For our video call use case, we look at audio capture, handling full duplex situations, and audio playback. Quality audio capture provides good voice intelligibility, a good signal-to-noise ratio (SNR), satisfactory directivity, and good management of audio when the user interacts with the laptop (such as typing during a call.)
The volume attribute covers the loudness of both capture and playback (measured objectively), as well as the ability to render both quiet and loud sonic material without defects (evaluated both objectively and perceptually).
A laptop needs to provide users with good readability, no matter the lighting conditions. Its color rendition should be faithful in the SDR color space (and in the HDR color space for HDR-capable devices. Main challenges:
We evaluate the electro-optical transfer function (EOTF), which represents the rendering of details in dark tones, midtones, and highlights. It should be as close as possible to that of the target reference screen but should adapt to bright lighting conditions to ensure that the content is still enjoyable.
Exposure measures how well the camera adjusts to and captures the brightness of the subject and the background. It relates as much to the correct lighting level of the picture as to the resulting contrast. For this attribute, we also pay special attention to high dynamic range conditions, in which we check the ability of the camera to capture detail from the brightest to the darkest portions of a scene.
The color attribute is a measure of how faithfully the camera reproduces color under a variety of lighting conditions and how pleasing its color rendering is to viewers. As with exposure, good color is important to nearly everyone. Pictures of people benefit greatly from natural and pleasant skin-tone representation.
The texture attribute focuses on how well the camera can preserve small details. This has become especially important because camera vendors have introduced noise reduction techniques that sometimes lower the amount of detail or add motion blur. For some applications, such as videoconferencing in low-bandwidth network conditions, the preservation of tiny details is not essential. But users using their webcam in high-end videoconferencing applications with decent bandwidth will appreciate a good texture performance score.
Texture and noise are two sides of the same coin: improving one often leads to degrading the other. The noise attribute indicates the amount of noise in the overall camera experience. Noise comes from the light of the scene itself, but also from the sensor and the electronics of the camera. In low light, the amount of noise in an image increases rapidly. Some cameras increase the integration time, but poor stability or post-processing can produce images with blurred rendering or loss of texture. Image overprocessing for noise reduction also tends to decrease detail and smooth out the texture of the image.
90f70e40cf