Artificial Intelligence 5 min read

Kuaishou Showcases AI‑Driven Multimedia Innovations at China Multimedia 2024

At the China Multimedia 2024 conference in Yinchuan, Kuaishou presented its latest AI‑driven large‑model technologies—including text‑to‑image, text‑to‑video, and audio models—alongside advances in intelligent video coding, a new research‑fund initiative, and recent industry awards.

Kuaishou Tech
Kuaishou Tech
Kuaishou Tech
Kuaishou Showcases AI‑Driven Multimedia Innovations at China Multimedia 2024

Recently, the China Multimedia 2024 conference was held in Yinchuan, Ningxia, co‑organized by the Chinese Society of Image and Graphics (CSIG) and the China Computer Federation (CCF). Kuaishou attended and shared its frontier AI‑driven multimedia technologies and application cases.

During the event, Kuaishou Vice President and Head of the Large‑Model Team, Zhang Di, delivered a keynote titled “KeLing AI Visual Generation Large Model and Applications”. He highlighted Kuaishou’s ongoing breakthroughs in foundational models, including the Kuaishou large language model, text‑to‑image generation model, text‑to‑video generation model, and speech model.

Zhang explained Kuaishou’s three‑pronged large‑model application strategy—enhanced understanding, innovative interaction, and generative exploration—aimed at creating next‑generation AIGC creation tools, lowering barriers for intelligent content production, and upgrading content understanding and distribution to improve user experience.

In the “Future Visual Information Coding and Intelligent Technology” forum, audio‑video algorithm expert Gu Jiwen presented “Frontier Exploration and Forms of Intelligent Coding and Decoding”. He discussed the rapid development of deep‑learning‑based video compression, the challenges of high on‑device compute costs, and Kuaishou’s solutions such as KVQ, Kuaishou KEP/KRP enhancements, and the KVC encoder.

Gu further described end‑to‑end optimizations across the audio‑video pipeline, including kernel‑level algorithm improvements and potential future directions for intelligent coding and decoding.

Kuashou also emphasized its commitment to building a robust AI‑industry‑academia‑research ecosystem through collaborations with societies and top universities, and announced the joint establishment of the “CCF‑Kuaishou Large‑Model Explorer Fund” during the 2024 World AI Conference. The fund will support up to 12 research projects with grants of up to 300,000 CNY each, focusing on next‑generation large‑model key technologies.

The company’s technical achievements were recognized with the “China Multimedia Enterprise Innovation Technology Award” for its video coding technology (KVC) and the “China Multimedia Outstanding Service Contribution Enterprise” award.

Looking ahead, Kuaishou plans to continue investing heavily in AI, accelerating technology innovation, promoting research‑to‑industry conversion, and fostering a thriving multimedia technology ecosystem together with partners.

AIlarge modelsmultimediaVideo CodingKuaishouresearch-fund
Kuaishou Tech
Written by

Kuaishou Tech

Official Kuaishou tech account, providing real-time updates on the latest Kuaishou technology practices.

0 followers
Reader feedback

How this landed with the community

login Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.