WAN-VACE Dual-Reference

This is a dual-image reference video generation application based on Wan VACE.

Upload two reference images and enter a motion description. The application will automatically identify and extract the subject information from both images, blend them into a video reference condition, and generate a new dynamic short video. It is best suited for effects like "subject from image 1 combined with weapons / props / elements from image 2 to generate a new video."

Recommended usage:

• Choose clear reference images with a single subject and clean backgrounds for both pictures as much as possible.
• Image 1 is better suited as the main character / subject image, while Image 2 is better suited as the weapon / prop / additional element image.
• The styles of the two images should not differ too much; otherwise, the fusion result may easily become unstable.
• Try to clearly describe in the prompt "what action the subject in image 1 performs and how the elements in image 2 participate in the scene."
It is more suitable for short-to-medium videos; starting with overly complex, large-scale scene changes is not recommended.

Recommended input example:
The subject in image 1 wields the weapon in image 2 with powerful movements, a slow camera push-in, clear details, natural lighting and shadows, and a unified overall composition.

🎁Claim RH Coins first, then try the workflow! Click your avatar in the upper right corner → Invitation Code → Enter 【rh-v1111】 to instantly receive 1000 RH Coins. Log in daily to claim another 100 coins~ For more ComfyUI workflows, tutorials, and gameplay, follow our official WeChat account (AIKSK), with simultaneous updates on TikTok / Bilibili / Xiaohongshu / YouTube (AI-KSK) ✨