Wan-alpha plus plus generates controllable transparent video effects
Wan-Alpha++: Animating a Single RGBA Asset into Controllable Transparent Video Effects
Computer Vision and Pattern Recognition
Summary
Creating videos with transparent parts is important for movies and games, but it’s hard to keep the see-through parts smooth and realistic over time. The authors made a new method called Wan-Alpha++ that helps generate videos with clear transparent regions and lets creators control how these look. Their technique works with both text commands and images as starting points and improves how transparent areas appear without causing flickering or disappearing effects. This makes it easier to produce high-quality visual effects that blend well with other content.
What this means in practice
- •For visual effects artists: Generate controllable transparent video assets for seamless compositing in films and games.
- •For graphic designers: Create animated transparent visual elements from text or images for digital content production.
Authors
Haotian Dong, Xin Wang, Wenjing Wang, Chen Li, Jing Lyu, Yan Tina Luximon, Ping Li, Wanchao Su, Ke Xu, Di Lin
Abstract
RGBA video generation is essential for creating compositable visual assets in filmmaking, game development, and digital content production. However, generating high-quality RGB appearance alongside temporally coherent transparency remains a significant challenge. Existing text-to-RGBA video methods often entangle RGB and alpha distributions and offer limited control over the precise appearance of a target asset. In addition, incorporating image conditioning typically suffers from the emergence-disappearance effects, which fail existing first-frame or first-and-last-frame conditions due to their fully transparent boundary frames. To address these issues, we present Wan-Alpha++, a unified framework that brings controllable RGBA video generation to both text- and image-to-video diffusion backbones through a shared RGBA latent space, extending our conference version Wan-Alpha from the text-conditioned setting to image-conditioned generation with an interior reference frame. Our method first learns shiftable RGB-alpha distributions while preserving the pretrained RGB prior. Specifically, we introduce a transparency-aware bidirectional diffusion objective to produce distinguishable RGBA latent representations, and a transparency-guided mean sampler that shifts the noise distribution to improve transparency generation and enable spatial controllability. Building upon this RGBA generation prior, we introduce temporal phase reparameterization for image-conditioned effect generation. By cyclically transforming a zero-positive-zero alpha pattern into a positive-zero-positive pattern, it relocates visible content to both temporal boundaries, facilitating effective appearance conditioning with limited RGBA data.