ShengShu Technology Launches Vidu S2 with Real-Time AI Video Interaction and Editing
Event summary
- ShengShu Technology unveiled Vidu S2 on September 15, 2026, featuring two models: S2-Avatar for real-time interactive digital characters and S2-Editing for real-time video stream editing.
- S2-Avatar supports dynamic reference image updates during interaction, improving resolution from 540p to 720p and enhancing instruction following for complex movements.
- S2-Editing enables real-time style transfer, outfit changes, subject replacement, and background replacement in incoming video streams.
- The technology includes spatial video capabilities for VR headsets, though this remains at an exploratory stage.
- Development was led by Jintao Zhang, a PhD student advised by Professor Jun Zhu.
The big picture
ShengShu Technology's Vidu S2 represents a significant step forward in AI-driven real-time video interaction and editing, addressing the growing demand for dynamic, high-resolution content in both digital and VR environments. The launch underscores the company's focus on pushing the boundaries of streaming video generation, potentially setting new standards for real-time media manipulation. The integration of spatial video capabilities, though still in exploratory stages, highlights the broader industry trend toward immersive digital experiences.
What we're watching
- Technical Execution
- Whether ShengShu Technology can sustain the performance improvements in real-time video generation and editing as it scales.
- Market Adoption
- The pace at which VR headset manufacturers and content creators integrate Vidu S2's spatial video capabilities.
- Competitive Positioning
- How Vidu S2's real-time editing and interaction features will differentiate it from existing AI video tools in the market.
Related topics