H.265, which produces videos of the same quality as H.HEV264 yet taking up only half the bandwidth, is poised to replace its predecessor as the mainstream video coding standard. Full-fledge deployment in the surveillance industry is inevitable, even though major challenges, such as complexity of coding and an imbalance between cost and functionality, still need to be overcome.
H.265, which produces videos of the same quality as H.264 yet taking up only half the bandwidth, is poised to replace its predecessor as the mainstream video coding standard. Full-fledge deployment in the surveillance industry is inevitable, even though major challenges, such as complexity of coding and an imbalance between cost and functionality, still need to be overcome.
H.265 is the next-generation video compression format that claims to be twice as efficient as H.264, the current industry standard. In Q4 2013, the first milestone in a continuous research effort to apply this new compression technology in broadcast was achieved. However, for the security and surveillance industry, H.265 is still a myth and not yet implemented on a wide scale.
Since HD is the ultimate trend, H.265 will be crucial in getting 4K “Ultra HD” contents to our televisions, PCs and tablets over the next few years. Achieving those objectives depends on the core components of the H.265 codec that are being developed and will probably launch in 2014.
H.265 is a video compression format that delivers crisper network video contents and requires 50% less bandwidth for the same image quality as H.264, making it easier to download or stream HD video. That said, over the next few years we can view Full HD contents online with our mobile devices such as cell phones and tablets.
H.265 supports 4K (4096×2160) and 8K (8192×4320) Ultra HD (UHD) contents as well. This makes H.265 an ideal technology for delivering high-quality IP videos and creates the possibility that H.265-supported IP cameras, DVRs, and NVRs will roll out over the next few months.
The H.264 standard has dominated the market for the past five years. Now, with H.265 set to become the mainstream, we must understand what the advantages of H.265 and their best-fit applications are. The following article explains the H.265 codec and its future trends.
WHAT IS H.265?
H.265 is a video compression standard, a successor to H.264, that was jointly developed by the ITU-T Video Coding Experts Group (VCEG) and related groups.
On August 22, 2012, Ericsson announced the world's first H.265 encoder, the Ericsson SVP 5500. In only six months, H.265 (also known as High Efficiency Video Coding, HEVC) was approved as the official successor to H.264 as an ITU-T standard.
H.265 is said to double the data compression ratio compared to H.264 at the same level of video quality. It can alternatively be used to provide substantially improved video quality at the same bit rate. It can support 8K UHD and resolutions up to 8192x4320.
H.265 was designed to significantly increase coding efficiency compared to H.264 – that is, to reduce bit rate requirements by half with comparable image quality. Depending on the application requirements, H.265 encoders can mitigate computational complexity; improve compression rate, robustness and error correction; and reduce encoding latency. While H.264 can transfer SD contents in 1Mbps, H.265 is able to transfer 720p (1280x720) and HD content in 1-2 Mbps.
Some leading IT pioneers like Qualcomm, Broadcom and Huawei have been showcasing related products enabled with H.265, which many believe is likely to replace H.264 as the global major compression standard.
WHY IS H.265 SUPERIORTHAN H.264?
H.265/HEVC video coding uses the same approach as H.264/AVC, including inter-/ intra-picture prediction, transform coding, quantization, deblocking filter, and entropy coding. But H.265/HEVC comprises coding units (CUs), predict units (PUs) and transform units (TUs).
Compared to H.264/AV, H.265/HEVC provides different methods to reduce the compression rate. Each marcoblock in H.264/AVC is 16x16 pixels, while in H.265/ HEVC the marcoblock provides different options, from 8x8, 16x16 to 64x64 pixels. What H.265 does is analyzing video contents and breaking them down into CUs of different sizes. For example, it would take much smaller blocks (down to 4x4 pixels) to encode detailed areas, like a vehicle in a parking lot, and much bigger blocks to encode the background, which contains less video data.
H.265 employs 33 directional modes for intra-prediction, compared to the 8 directional modes for intra-prediction used by H.264. With the same image quality, H.265 delivers 39 to 44% less decoding bit streams than H.264. This result may vary with different quality assurance methods.
Peak signal-to-noise ratio (PSNR) and subjective assessment of video quality are standard ways to measure the coding efficiency of a video coding mechanism. In a subjective assessment of video quality, H.265 delivers the same or better coding efficiency than H.264 with bit streams reduced by 51 to 74%. This result is very important since subjective assessment of video quality is considered to be the most important way to measure a video coding standard, as humans perceive video quality subjectively.
Like H.264, which uses the Main, Baseline and Simple profiles, the first approved version of the HEVC/H.265 standard includes Main, Main 10, and Main Still Picture profiles.
The Main profile supports bit depth of 8 bits per sample, which allows for 256 shades per primary color, translating into a total of 16.8 million colors. The Main 10 profile supports bit depth of 10 bits per sample, which allows for 1024 shades per primary color or a total of 1.1 billion colors, making it ideal for UHDTV. They both use 4:2:0 chroma sampling.
HEVC also contains provisions for additional profiles. Future extensions that are being discussed for HEVC include increased bit depth, 4:2:2 and 4:4:4 chroma sampling, multiview video coding (MVC), and scalable video coding (SVC). On January 8, 2013, Broadcom announced in CES the BCM7445, which is an Ultra HD decoding chip capable of decoding HEVC of up to 4096x2160p at 60 fps. The BCM7445 is a 28-nm ARM-architecture chip capable of 21,000 Dhrystone MIPS with volume production estimated for the middle of 2014.
FUTURE TRENDS OF H.265 CODING
Security and surveillance vendors are more concerned about the time-tomarket of devices that support H.265/ HEVC coding. This will depend on many factors -- for example, how soon will graphics chips makers, such as AMD and NVIDIA, integrate H.265 into their products. The first generation of H.265 chipsets may be only suitable for initial video coding standards, and there are still many H.265 image processing functions, extensions, and multiview video coding that need to be enhanced.
In 2013 CPSE, Hisilicon and Grain Media did a demo on the initial results of their H.265 research. All of the invited visitors were impressed and agreed that H.265 had a huge potential. However, challenges still remain. These include H.265 coding complexity and an imbalance between cost and product features.
Many manufacturers in the security and surveillance industry hope that with the efforts of chipset and IC technology vendors, HEVC will quickly replace H.264 as the dominant IP surveillance coding technology. While most cable TV and digital TV broadcasters are still using the MPEG-2 standard, the good news is that some TV broadcasting companies have replaced MPEG-2 with H.265 standard for HDTV applications, as H.265 reduces bandwidth consumption by 70 to 80% and can support full HD 1080P TV broadcasting with existing bandwidth conditions. Compared to cable TV companies, satellite TV providers may adopt the H.265 standard earlier. The H.265 codec will be the ultimate solution for 4K and 8K options that can enable Ultra high definition television (UHDTV). But challenges on transmission, storage and playback still linger.
Right now, only few native 4K video contents are available, and manufacturers are just planning to launch 4K IP cameras in Q3 2014. The introduction of H.265 coding standard means that the theory is ready, but a unified method to transmit such ultra HD signals is not. Therefore, such development is quite crucial.
In addition, the storage of H.265 encoded videos would also be a problem. Even the Blu-ray Disc Association is working to find a solution to enable storage of 4K videos on Blu-ray discs. Theoretically, this should become possible with an extension of the H.264 format, but bit streams become an issue. Only discs with at least 100GB capacity will enable storage of Blu-ray 4K movies encoded with H.264. But where can you find 100GB re-writable discs? In other words, even though H.265 coding and chipset components are ready, it still lacks the storage and playback solutions that support 4K contents and that are compatible with the existing Blu-ray Disc standard. This is the major challenge for H.265 development.
Since H.265 is the future of television and video, will it become the mainstream for security and surveillance as well? Professional surveillance vendors are not sure either, since H.265 deployment in security and surveillance is not only subject to the challenges mentioned above but also dependent on its end-users.
There are project-based professional users and general consumers for surveillance applications. Professional users are those who conduct city surveillance, traffic monitoring and bank monitoring and who demand solutions that are more stable and reliable. Most of them already have existing technologies and therefore would be hesitant to adopt H.265, which requires a longer verification period.
On the other hand, SMB users and consumers such as home and shop users are faced with less installation cost and are therefore more likely to adopt the new technology. For this reason, H.265 could have its early success in SMB applications and gain acceptance in the consumer market.
With the wide adoption of IP surveillance, HD trends are everywhere. Yet at the same time, limitations of the H.264 standard have begun to emerge. 1080P video contents, for example, require 4 to 10Mbps bandwidth, creating huge cost pressure for service providers and SIs. Network bandwidth and storage equipment usually account for 40 to 50%of system investments in the surveillance ecosystem, and bandwidth cost and maturity of equipment technology are major issues facing operators offering Full HD.
If H.265/HEVC standard matures quickly, and its compression efficiency is improved by 50% over H.264, it can result in 20% investment savings, ensuring higher performance and lower network and system building cost in video surveillance.
H.265 has more superior features than H.264, so it's just a matter of time before H.265 standard and components are ready for market. We are cautiously optimistic about the future of the H.265/HEVC standard, as HD 1080P TV broadcasting and 4K video streaming with H.265 are already possible. There is even the possibility that mobile devices will be required to support H.265, and manufacturers will try their best to accommodate.