Skip to main content

Five ways video coding will evolve over the next five years

Woman with VR headset

The landscape for video coding – the process of compressing and decompressing video files - is changing to enable highly scalable distribution for entertainment, social media, and communications. Video is becoming ever more pervasive, and the dawn of a new era for artificial intelligence presents challenges and opportunities not seen in a generation. At Nokia, we have been pioneers in audio-visual coding for decades and continue to play a key role in the evolution of the field. That means we spend a lot of time thinking about what the future holds. Here are five major trends that we expect to see in the next five years:

1. Video will be used & consumed in more diverse ways

We’re not just talking about switching from monitors to head-mounted displays. It also involves video being processed and interpreted by AI applications within everything from autonomous vehicles to smart devices. A change in consumption means tools such as perceptual quantization (based on the human visual system) may not always be appropriate and the way we evaluate video codecs may change. Holographic rendition and augmented reality will become more commonplace, and there will be a need to blend captured and synthetic content seamlessly. The countless new use cases will require codecs capable of handling such diverse needs efficiently and qualitatively.

2. The way we design codecs will continue to change.

AI is already making its presence known in the area of video coding, and it will continue iteratively on this journey. Encoders are ‘non-normative’, meaning that the way they operate is not specified by the standard. Manufacturers have substantial freedom to innovate, and they are already seeing performance benefits from the inclusion of AI tools in their products.   Researchers developing ‘normative’ video codec algorithms are now utilizing AI tools—effectively a smarter algorithm design process—and this will continue to take hold over the coming years. Looking further into the future, we can envisage a time when the codec itself may self-optimize and reconfigure itself in real-time. For example, drone surveillance could self-adjust the codec’s design parameters to suit the terrain it is covering. This will require continued careful research (with an eye on what can feasibly be implemented in available hardware), but it’s not unthinkable that it could be achieved before the decade is out.

3. Coding efficiency remains a key driver

Any opportunity to reduce video bitrate frees up storage and provides broadcasters with greater freedom in how they present their content to customers. While computing power and storage capacity increase over time, the amount of video content is increasing at an even greater pace, therefore we expect service providers to continue valuing coding efficiency.  That said, it has historically been the case that new applications and features incentivize adoption of a new codec, HDR being one example. Moving forward, the traditional requirements of efficiency and low bitrate need to be balanced with the ever-increasing demand from consumers for richer, more engaging video content.

4. Competition will be balanced with realism

Competition between video codecs is increasing as various rival options emerge. We believe that codecs produced in a transparent and collaborative fashion serve both the public and the industry well. This process has historically brought together the best work from researchers across the globe and produced top-quality results. However, competition means we can no longer assume that standards bodies’ codecs will be unanimously adopted. This, along with economic pressures, may also lead to more grounded expectations of what can be achieved in the standardization sphere.

5. Disruption could be on the horizon

Versatile Video Coding, the latest video coding standard, continues to be adopted by an increasing number of hardware applications and is likely to surpass older codecs in the coming years. But technology evolves, and explorations are already being made into how new standards can provide further coding efficiency gains. Most codecs have used the same basic structure or ‘block diagram’ for at least two decades. A lot has been invested in tuning and optimizing this core design, but at some point—possibly before 2030— it will be challenged by a disruptive new approach to video. Nokia is fully invested in exploring new ways to help drive new innovation, efficiency and enhanced user experiences including the possibilities of new codecs in the future.

If you want to get involved in discussing this topic, then you have the chance to discuss it with me in person this month. I’ll be taking part in a panel discussion entitled “Advanced Video Coding: the Future Outlook to the Next 5 Years” at the ACM Mile High Video conference in Denver, Colorado that runs 11th-14th February 2024. Let’s shape the future of video together.


Justin Ridge

About Justin Ridge

Justin has been with Nokia for over 20 years and is currently a Principal Engineer in Nokia Technologies’ Video Coding Research group.  He’s been involved in video codec standardization since H.264/AVC days and is currently a vice-chair of ITU-T Study Group 16, and President of the Media Coding Industry Forum (MC-IF).  He’s based in Dallas (where he has a habit of rescuing cairn terriers) but you’ll detect an Australian accent if you talk with him.

Connect with Justin On LinkedIn

Article tags