Three Reference Images Plus One Prompt: High-Motion Fight Scenes in Seedance 2.5
techhalla · x · 2026-10-21
The author lays out a full workflow: generate three Midjourney reference images (one environment, two characters, prompts in the image ALT text), then feed them into Seedance 2.5 with a long prompt specifying hyper-kinetic non-stop combat, overcast misty alpine daylight, handheld 35mm with heavy motion blur, whip pans and hard cuts every 1-2 seconds, crash zooms, and debris/water-spray particle systems, using <<image1>> as the exact environment reference (grassy alpine meadow, limestone boulders, cliff waterfalls, floating islands).
More from Multimodal
- Higgsfield demo imagines a coin-sized black hole appearing right next to you — mhdfaran · 2026-10-21
- Cat vs Kaiju: an AI-generated monster-movie short worth watching — Reasonable_Damage_18 · 2026-10-21
- Rethinking vanilla pixel diffusion transformers: capacity limits and a high-frequency deficit in JiT — CSProfKGD · 2026-10-21
- Stunning gothic AI-generated image: skull throne, black panthers, ornate detail — tetsuoai · 2026-10-21
- Virtual British rocker built with Suno duets and Dreamina music videos — AIandDesign · 2026-10-21
- Cinematic 'What If You Stole $10M' Animation Made with Higgsfield Supercomputer — EXM7777 · 2026-10-21