Vidu Launches Q2 “Reference-to-Video”
With higher consistency, sooner technology, and extra inexpensive pricing, ShengShu Technology gives a top-tier different to main world platforms.
ShengShu Technology, a world chief in multimodal generative AI, in the present day launched Vidu Q2 “Reference-to-Video”, setting a brand new commonplace for AIGC merchandise. This launch shifts AI from merely creating movement to performing it, mixing know-how with creativity to ship lifelike, emotional, and cinematic movies.
The new “Reference-to-Video” characteristic permits creators to make constant, expressive movies utilizing as much as seven reference photos for faces, gestures, scenes, or props. It can mix a number of unrelated parts, like totally different characters, gadgets, or backgrounds, right into a single, unified video. With a textual content immediate, it blends these parts utilizing its Multiple-Entity Consistency characteristic, protecting every half distinct and true to its unique look, even in advanced or altering scenes. With sooner technology and extra inexpensive pricing, Vidu Q2 “Reference-to-Video” now competes with one of the best video platforms, bringing the “AI efficiency period” nearer.
“Vidu Q2 ‘Reference-to-Video’ marks a brand new chapter in AI video creation,” mentioned Yihang Luo, CEO of ShengShu Technology. “We’re transferring right into a time the place AI can mimic human appears and specific feelings with cinematic aptitude. This launch goes past fundamental video creation; it’s about instructing AI to behave and inform tales alongside creators.”
Bringing Realism and Cinematic Quality to AI Video
Vidu Q2 “Reference-to-Video” delivers a major leap in realism, capturing refined feelings like a hesitant smile, a curious look, or tense anticipation with pure stream. Movements really feel clean and alive, changing stiff, robotic motions with vibrant vitality.
The platform additionally handles cinematic methods like clean digicam shifts, panning, and depth of subject, mimicking skilled filmmaking. Scenes transition seamlessly between large photographs and close-ups, letting creators craft tales with each grandeur and emotional depth.
Vidu Q2 “Reference-to-Video” additionally higher understands prompts, capturing the meant temper and that means extra precisely. This makes it simpler for creators to deal with storytelling reasonably than fixing errors, turning generative video right into a sensible, on a regular basis software.
With sooner technology and better accuracy, creators can produce longer, expressive movies for movie, animation, promoting, and industrial content material with minimal changes.
Swift Market Adoption Proves Real-World Value
With the launch of Vidu Q2 “Reference-to-Video”, the Vidu Q2 MaaS API was additionally made obtainable globally, enabling companies to simply add Reference-to-Video options to their work processes. Building on its established experience in Reference Generation know-how, ShengShu rapidly shaped a lot of new partnerships with promoting and e-commerce corporations, demonstrating the mannequin’s high-quality output and ease of use. The replace helps companies by offering new methods to scale back prices, enhance effectivity, and improve artistic high quality.
The mannequin’s excessive consistency is very priceless for industrial video manufacturing. Even with advanced digicam actions or character interactions, the principle topic and product particulars stay clear and secure, creating lifelike 360-degree shows that look identical to actual footage. In product promoting, fashions now show pure gestures and micro-expressions, producing extremely lifelike and fascinating footage that redefines what AI-generated content material can obtain.
A Continuing Story of Innovation
Since its founding, ShengShu Technology has led the way in which in artistic AI with groundbreaking advances. In 2022, the workforce launched the U-ViT structure, the world’s first Diffusion–Transformer hybrid mannequin, earlier than the DiT structure utilized by high opponents. The firm additionally created UniDiffuser, the primary mannequin to generate each textual content and pictures in a single system.
Building on these, ShengShu’s Analytic-DPM framework made AI processing as much as 20 occasions sooner, incomes the ICLR 2022 Outstanding Paper Award. Its DPM-Solver turned the world’s quickest diffusion solver, later utilized in platforms like Stable Diffusion.
Each step within the Vidu collection turned these breakthroughs into artistic platforms: Vidu 1.5 introduced the world’s first constant multi-character scenes, Vidu 2.0 provided 10-second movies at half the business price, and Vidu Q1 added cinematic transitions with lifelike sound. Now, Vidu Q2 “Reference-to-Video” combines these advances into one highly effective, expressive system, transferring the platform from AI creation to true AI efficiency.
A Rapid Rise to Global Top Player
Since Vidu’s launch in April 2024, ShengShu Technology has seen fast development. In only one 12 months, the platform reached over 200 nations, gained 30 million customers, and produced over 400 million movies, cementing its place as a high participant in artistic AI.
“With every launch, we mix know-how and creativity extra intently,” mentioned Luo. “Our purpose isn’t to interchange creativity however to broaden it, making creativeness seen and feelings limitless.”
The submit Vidu Launches Q2 “Reference-to-Video” first appeared on AI-Tech Park.
