Visionular at IBC 2026: AI and Video in Practice

From September 11–14, Visionular joined customers, partners and industry professionals in Amsterdam for IBC 2026. Across four days of demonstrations and discussions, we explored a shared challenge: how to meet growing demands on video infrastructure while preserving quality and improving efficiency.

At our booth and on stage, we shared Visionular’s approach to video infrastructure built for humans and AI, from efficient encoding and visual enhancement to optimising AI-generated content.

Our booth gave visitors the opportunity to explore Visionular’s technology and discuss its application across broadcast, streaming and AI workflows.

Video compression remained a central part of those conversations. Our Aurora encoder family — Aurora4 for H.264/AVC, Aurora5 for H.265/HEVC and Aurora1 for AV1 — combines AI preprocessing, content-adaptive encoding and intelligent bitrate control to improve compression efficiency across live, on-demand and real-time video.

Discussions focused on what this means in practice: reducing bandwidth and storage requirements, managing encoding resources and preserving visual quality. With SDK, AuroraFlex, and AuroraCloud deployment options, teams can evaluate these capabilities within their existing infrastructure.

Each conversation started with the practical needs of the teams we met: the content they handle, the platforms they operate and the challenges they face as their services grow.

GenOpt works alongside generative AI models to improve clarity, fine detail, realism and consistency between frames. It addresses common visual issues such as soft textures, unnatural skin and hair, jagged edges and flickering. By enhancing generated output, it can also reduce reliance on higher-cost generation settings and repeated generation attempts.

These capabilities complement our work in video enhancement. AuroraEnhancer provides tools for super-resolution, denoising, detail restoration, colour processing and frame interpolation, helping teams improve source material across recorded and live workflows.

The discussions also reflected a broader shift: video increasingly serves both human audiences and AI systems that analyse and interpret content. Supporting these different uses creates new requirements for how video is encoded, delivered and processed.

IBC provided a valuable setting to explore those requirements with the people building and operating today’s video services.

Visionular’s participation extended to the conference programme, where Zoe Liu, CEO and co-founder, discussed real-time video and the industry’s evolving technology needs.

Her presentation explored how AuroraVision supports live sports production through software. Capabilities include automatic subject tracking and framing, slow-motion generation and super-resolution for 4K and 8K output. These functions can be integrated into workflows using existing cameras, GPU servers and distribution systems, with modules selected according to production requirements.

An interview with Zoe and editorial coverage of Visionular’s technology in the IBC Daily offered further opportunities to share our perspective with the show’s audience.

Together, these activities connected our technology with practical questions about production quality, infrastructure and the resources available to different teams.

We left Amsterdam with new connections and discussions to take forward. Our focus now is on translating the requirements shared during the show into technical evaluations, demonstrations and opportunities for collaboration.

Whether the next step is benchmarking an encoder, testing enhancement on existing footage or evaluating GenOpt within an AI generation workflow, the starting point is the same: understanding the content, the quality targets and the operational constraints.

Thank you to everyone who visited our booth, attended the sessions and took the time to share their challenges with us.