Fusing 3JS Drafts with Image and Video Gen for Precise AI Video Control
mattshumer_ · x · 2026-08-12
Developer Matt Shumer shared a new workflow to solve the lack of control in AI video generation. The technique involves drafting motion in 3JS, generating the visual style with an image generation model, and fusing them with a video generation model. Building upon an idea by gmicloud, Shumer improved the approach to increase stability and allow for flexible style control, fixing issues where the original style barely changed.
Related event: ThreeJS Combined with Video Models Achieves Precise AI Video Control(6 posts)→
More from Multimodal
- RTX 5090 Test: Generate MiniMax H3 Videos in Just 5-6 Minutes — tekprodfx16 · 2026-08-13
- FLUX 3 Stunningly Generates Parkouring Robot Escaping Police Chase — Dr_Singularity · 2026-08-13
- New Trick: Using LTX2.5 to Upscale MiniMax H3 Videos — smereces · 2026-08-13
- Comparison: Impact of LoRA Strength and SageAttention on Video Generation — Mysterious_Pride_858 · 2026-08-13
- Minimax H3 Video Generation Test: 25 Steps is the Lowest Recommended Setting — rm_rf_all_files · 2026-08-13
- Open-Sourcing Signet Trainer: Multimodal LoRA Training for Minimax H3 and LTX — lmofr · 2026-08-13