Luis Ochoa and Francisco Utray, April 5, 2016.

Technical solutions for High Dynamic Range

 

In this second part of the HDR guide we will take a brief look at the solutions being put forward by industry and standardization bodies for high dynamic range and gamma curves for Ultra HD-HDR image production and display.

The challenge is to increase the maximum brightness level and at the same time lower the black level. Progress is being made in both directions. The new high dynamic range (HDR) displays will be able to present images with more detail in highlights and deep shadows. The goal is, above all, to achieve more rendering power in the highlights. The higher the dynamic range of a monitor, the more detail it will be able to display in the highlights.

Another important thing to keep in mind is that not all images require these levels of dynamic range. If the scene being depicted does not have very high light contrast ratios, it will not be necessary to take advantage of the full capabilities of an HDR monitor. You will notice the difference only in those shots that require it. For example, a scene shot indoors with controlled lighting does not need the same level of contrast as an outdoor backlight on a sunny day.

In acquisition, in order to achieve 14 or 15 f-stops of dynamic range, camera manufacturers use a logarithmic gamma curve. There is no standard, each manufacturer has its own logarithmic gamma curve (Arri, Red, Sony, Canon, Panasonic, etc.), although all of them are more or less based on Kodak’s CINEON curve. To view these images correctly on today’s monitors, a conversion is made – by means of a LUT or a color correction process – to adapt the signal to the BT-709 standard. But at that point, we are already in the BT-709 environment, and the high dynamic range and Wide Color Gamut (WCG) have been lost.

It is therefore necessary that a gamma correction curve be standardized for the new HDR monitors and TVs so that camera manufacturers, colorists and mastering managers know exactly under what conditions they will be reproducing content.

 

Gamma uptake and monitoring curves

The gamma correction process in monitoring is called Electro-Optical Transfer Function (EOTF) and comes from the days of tube televisions (CTR). All current displays follow the BT-709 recommendations and apply a gamma correction of 2.4. A procedure with many limitations that no longer has a place in the world of Ultra High Definition and HDR.

An inverse curve, the Opto-Electronic Transfer Function (OETF), is used in acquisition. These are the gamma curves used by camera manufacturers to maximize the dynamic range in acquisition (log curves) or to directly generate processed images ready for broadcast (standard gamma curve BT-709).

 

Gamma correction in acquisition and in monitoring
EOTF	Electro-Optical Transfer Function	Gamma correction in monitoring. It is a function that assigns a brightness level on the monitor to each digital encoding value.
OETF	Opto-Electronic Transfer Function	Gamma curve that is applied in the camera. It is a function that assigns a digital encoding value to the different brightness levels of the scene that the camera is capturing.

Source: 709 MediaRoom (SMPTE 2015: 4).

 

The following graph compares the OETF proposals made by Dolby (Dolby-Vision) and the BBC (HLG HDR) with the BT-709 curve. It can be seen how the BT-709 curve breaks down in the highlights and how the new proposals correct this problem.

 

 

The issue for new HDR displays is to define an EOTF standard curve that takes full advantage of the high dynamic range rendering capability and leaves behind the limitations of the BT-709 standard (Borer 2014). This task is being addressed by international technical standardization organizations such as SMPTE, ISO or ITU among others.

The standardization of an EOTF curve to be integrated in all HDR displays is a fundamental prerequisite for the development of this new technology. All displays must behave the same regardless of brand and manufacturer. In this way, content providers and camera manufacturers can prepare the material for these devices.

Currently, for example, there is a Canon HDR field monitor. If we are recording with a Canon camera, the monitor will be able to present the images in high dynamic range without the need for signal processing, since it is the same manufacturer who has designed the camera and the monitor. The standardization of an EOTF curve will allow images from any camera to be viewed in HDR on any display. The standardization of an EOTF curve will also make it possible to establish the criteria for mastering HDR content for broadcast.

 

The SMPTE ST-2084 standard

The technical standard SMPTE ST-2084, for the moment, is winning the race to be the most accepted way to master, distribute and consume HDR content. Many manufacturers, such as Sony, Panasonic or Samsung, are adopting this standard for their new HDR displays.

SMPTE has adopted for this standard a part of Doby-Vision’s solution, “Dolby Perceptual Quantizer (PQ)”, which is offered as a royalty-free standard.

The standard proposes an EOFT curve standard for 10,000 nits displays that achieves increased detail in the highlights. This can be seen in the following image with the histograms of an SDR and HDR image. You can see the ‘clipping’ in the SDR signal at the 100 nits level and the detail in HDR up to 10,000 nits.

 

Hybrid Log-Gamma (HLG) (ARIB STD-B67)

Hybrid Log-Gamma (HLG) is a standard for high dynamic range jointly developed by the BBC and the Japanese NHK. It has been published in 2015 in ARIB, the Japanese technical standardization agency(Association of Radio Industries and Businesses)(ARIB STD-B67). HLG takes into consideration compatibility with standard dynamic range (SDR) displays and does not use metadata.

It proposes a different curve for the different maximum brightness levels of the displays with a limit at 5,000 nits. An interesting issue with this solution is that it takes into consideration the ambient brightness which will also affect the curve. In the following graph you can see an example of how the curve changes as a function of ambient light changes.

 

Conclusions

We have found in making this guide that HDR technology for television is still at an early stage of development.

The leap being pursued is the replacement of the current EOTF of the high-definition television standard (BT-709) with a new set of higher performance electro-optical transfer functions.

The large consumer electronics industry needs a new product, once the sales of ‘Full HD’ displays have stabilized. The baton has to be taken over by ‘Ultra HD’ which implies not only 4K, but also HDR and Wide Color Gamut (WCG). Let’s take a look at the expectations raised by Paul Erickson, Senior Analyst at IHS Technology.

“Ever-rising consumer adoption of Ultra HD TV sets will fuel strong growth for the entire Ultra HD ecosystem over the next few years. Annual worldwide shipments of Ultra HD TVs are expected to grow nearly 719% over the next several years according to IHS’ forecasts, from nearly 12 million in 2014 to nearly 96 million in 2019, with over 300 million in use by the end of 2019.” (UHD Alliance, 2016).

From the industry’s point of view, there seem to be unwaivable interests in pushing this new technology.

HDR a technology more oriented to television than to cinema projection in theaters. The increased brightness of the high lights on the big screen generates reflection in the room and therefore increases ambient light. Dobly-Cinema is a complete screening room design that attempts to alleviate this effect, preserving ‘dark room’ conditions but increasing dynamic range and peak brightness levels.

As image creation professionals, the question that arises is whether it is really necessary to make this leap and the acceptance that these new images will have in film and television audiences. We are facing the ‘never seen before’. Cinema, television, photography, painting, etc. have created the current aesthetic canons in our culture. The increase in the contrast ratios of representation may represent a paradigm shift from the perspective of artistic creation. It is not an easy challenge for cinematographers and colorists.

Early experience with Canon’s HDR mastering monitor is that subtle highlights on a character’s face can become a serious flaw that is difficult to correct on a brighter HDR screen. It is also true that general exterior shots with detail in the highlights will give superb images. This is the opinion of colorist Dado Valentic, an enthusiast of this new technology:

“It is refreshing to see that HDR is one of those technologies and innovations that won’t require any heavy marketing for it to be accepted by the end user. Once you glance at an HDR TV screen you won’t need any convincing – images look much better, fresher, and more exciting. If 3D was interesting for its novelty and the 4K advantages were not visible to all, it is clear to see that HDR is not going to fall into any of these two categories.” (Valentic 2016)

It will also require a process of adaptation on the part of the public, traditionally accustomed to the limitations of SDR, who will encounter new images that are ‘different’ from those that have been produced previously and which they may not necessarily like from the outset.

Finally, the health impact of viewing brighter images, on larger screens and from closer proximity, is worth considering.

Luis Ochoa and Francisco Utray, April 5, 2016.

 

References cited

ARIB (2015) “Essential parameter values for the Extended Image Dynamic Range Television (EIDRTV) system for programme production” ARIB STD-B67 Version 1.0. Available for download at http://www.arib.or.jp/english/html/overview/doc/2-STD-B67v1_0.pdf.

Blu-ray (2015), “White Paper Blu-ray Disc™ Read-Only Format (Ultra HD Blu-ray™). Audio Visual Application Format Specifications for BD-ROM Version 3.0”. Blu-ray Disc Association. Available for download at: http://www.blu-raydisc.com/assets/Downloadablefile/BD-ROM_Part3_V3.0_WhitePaper_150724.pdf

Borer, T. (2014) “Non-linear Opto-Electrical Transfer Functions for HDR TV” BBC White Paper WHP 283.

Barten, P.G.J. (2004) “Formula for the Contrast Sensitivity of the Human Eye”, Proc. SPIE-IS&T, 5294:231-238, Jan. 2004.

Geutskens, Yoeri (2016) “The State of #HDR in Broadcast and OTT – CES 2016 update” Available online at http://www.ctoic.net/blog/the-state-of-hdr-in-broadcast-and-ott-at-year-end-2015/

SMPTE (2015) “Study Group Report High-Dynamic-Range (HDR) Imaging Ecosystem” Society of Motion Picture and Television Engineers ®, Inc. (SMPTE ®).

UHD Alliance (2016) “UHD alliance defines premium home entertainment experience” Press release. Available online at: http://www.uhdalliance.org/uhd-alliance-press-releasejanuary-4-2016/#more-1227

Valentic, Dado (2016) “HDR SURVIVAL GUIDE – Next Generation Home Entertainment.” Available online at: https://www.linkedin.com/pulse/hdr-survival-guide-next-generation-home-dado-valentic