The bicycle capital of the world will once again welcome visitors from all over the world for this year's edition of IBC! Starting this Friday, the content and technology community will come together to gain the latest insights on inspiring content. Collabora will also be present to take part and share in this exciting event at the RAI Amsterdam. If you are planning on attending, please make sure to add booth B66 in Hall 5 to your itinerary & come say hello!

This year, we will be showcasing some of our recent work with Netflix, YouTube and Amazon Prime Video on the DAB protocol, as well as the software integration of LCEVC, MPEG's novel enhancement codec, into the GStreamer multimedia framework, and a machine learning video compression that runs on a complete open source stack.

• Device Automation Bus on the RDK

  Working together to improve the application testing and certification process for device partners, Netflix, YouTube, and Amazon Prime Video created the Device Automation Bus (DAB), an open-source network protocol that enables device manufacturers and application developers to remotely control a device under test. Collabora created a reference implemention of DAB, along with an example backend for the RDK platform. Drop by our booth to see DAB being used to control a RDK set top box including the Cobalt Open Source lightweight application container with the YouTube app.
  
  

• LCEVC meets GStreamer
  

  MPEG-5 Part 2 Low Complexity Enhancement Video Coding (LCEVC) is a new video standard. LCEVC works by encoding a lower resolution version of a source image using any existing codec such as H.264, H.265 or AV1 (the base codec) and the difference between the reconstructed lower resolution image and the source using a different compression method (the enhancement). This results in a greater compression ratio than the base codec alone. We'll be showing the GStreamer integration of LCEVC with the base layer being decoded by a hardware decoder using standard GStreamer plugins.
  
  

• Panfrost ML-based video compression
  

  Traditional video codec techniques are hitting their limit, newer codecs require a 10x increase in computational power for a 2x gain in coding efficiency. It has become clear that radically new approaches are needed. Many have suggested that Deep Learning techniques can yield better results within acceptable computational deadlines. This demo proposes such a technique specifically targeted at videoconferencing.

In addition to our booth, we will also be taking the stage on two separate occasions, a first for Collabora at IBC! Attendees will have the opportunity to learn about the multifaceted nature of GStreamer, but also the ability to video conference with limited bandwidth using machine learning.

Update: The video recordings for both presentations are now available, click on either talk listed below to start watching!

Collabora @ IBC 2023

GStreamer, the Open Source framework for broadcast
Presented by Olivier Crete - Friday, September 15
Content Everywhere Stage 1, Hall 5

The GStreamer Open Source multimedia framework has been used for years to power broadcast products and solutions. As the leading extensible multimedia framework, GStreamer provides everything needed. With its focus on streaming, GStreamer offers a WebRTC implementation perfected for over a decade but also supports all other protocols like SRT, MPEG-DASH, and DASH. GStreamer is not only the Swiss Army knife of streaming, it can also tie together various hardware accelerated solutions, from FPGA to GPUs to capture devices to dedicated media accelerators; it knows them all. We'll present the main tools that GStreamer offers the broadcasting world and how it can be used to easily create dynamic pipelines.

Open-source ultra-low bitrate compression for video conferencing without a camera
Presented by Marcus Edel - Monday, September 18
Content Everywhere Stage 2, Hall 5

Over the years, different video codecs have been developed to meet the needs of various broadcasting applications. The quality of these codecs is excellent at medium-to-low bitrates, but it degrades when operating at low bitrates. There has been a massive interest in replacing these methods with machine-learning approaches. Using open-source software, Collabora has developed a compression pipeline that enables a face video broadcasting system that achieves the same visual quality as H.264 while using a fraction of the bandwidth. Our pipeline uses a speech-to-text model to transcribe the audio feed. A generative text-to-speech model is used to recover audio from the text on the receiver side, followed by a lipsyncing model to reconstruct the face with the generated audio. This enables communication at lower bitrates in remote terrains with limited bandwidth, and frees bandwidth for error correction during broadcasting. We'll present the pipeline, and it's use-case for the broadcasting world.

As a leading contributor to the GStreamer, PipeWire, and Wayland projects, Collabora is uniquely positioned to provide the expertise your require to integrate the latest Open Source technologies with the newest multimedia standards. Be it for low latency streaming, cloud transcoding, embedded capture and playback or AR/VR, Collabora can help you with customization, optimization, hardware acceleration, and complex multimedia integration so you can build innovative solutions across the digital media space.

Please get in touch with us if you would like to meet at IBC. We look forward to seeing you at the RAI!

Related Posts

Pushing testing laboratory performance limits by benchmarking LAVA - Part 1

Radxa Rock-5B PCIe and RTL8125B networking in U-boot

Booting on Radxa's Rock-5B without any media used

Related Posts

Pushing testing laboratory performance limits by benchmarking LAVA - Part 1

Radxa Rock-5B PCIe and RTL8125B networking in U-boot

Booting on Radxa's Rock-5B without any media used