BMF - A Flexible and High-Performance Multimedia Framework

BMF (Babit Multimedia Framework) is a novel open-source multimedia framework developed by ByteDance and NVIDIA, born out of experience in serving over 2.5bn people that interact with ByteDance products including TikTok.. It has been widely used in video-related business scenarios in ByteDance, such as AIGC, short video, video streaming. Get introduced to it in this short video and how it can be used with AI in this article by our Evangelist Joshua Alphonse, or did right in to look at the code in GitHub. Want to dive deeper, come to this webinar, and learn from the experts.

BMF - A Flexible and High-Performance Multimedia Framework- Shaofei Wang
In this session we will demonstrate how BMF can achieve high performance with more efficient development cycles. We will discuss the challenges of deploying cloud-based products using FFmpeg and share our experience to overcome those challenges, which led to the design of BMF. We will discuss its key features, including ease of use, high performance, and full compatibility with FFmpeg.

We will also dive deep into the details of how BMF and GPU could benefit the generative AI workloads, such as the optimized pipeline for video transcoding + AIGC inference. Next, we will demonstrate how efficiently BMF could help everyone, from prototyping a fascinating AIGC feature with just a few codes to deploying a GPU-optimized product in the data center.

How to Create an AI video similarity Search with BMF and Pinecone - Joshua Alphonse
Video similarity search or reverse video search works like reverse image search but uses a video to find other videos that are alike. Essentially, you use a video to look for matching ones. While handling videos is generally more complex and the accuracy might not be as good as with other models, the use of AI for video tasks is growing. In this talk, Joshua will show you how to create a reverse video search application from scratch.

About The Speakers：
Wang, Shaofei
Shaofei has worked in the multimedia industry for over 10 years. An architect in the media infrastructure team of Bytedance's Video Architecture Department. His work mainly focuses on multimedia framework and infrastructure to gain acceleration of video processing, strong heterogeneous computing capabilities and cross team cooperation on multimedia application requirements.

Joshua Alphonse is an award-winning Developer Advocate, speaker, and engineer from New York. As a Developer Advocate for ByteDance Open Source Program Office (OSPO). Previously, Joshua worked at Wix, leading Product and R&D engagements for their Developer Relations Team. He successfully created content, tutorials, and curated events for the web development community.

About Bytedance: ByteDance is a global incubator of platforms at the cutting edge of commerce, content, entertainment and enterprise services - over 2.5bn people interact with ByteDance products including TikTok.