Keeping character scale consistent in MiniMax H3 Ref2V: prompting tricks that mostly fail
fluvialcrunchy · reddit · 2026-09-16
A creator working with MiniMax H3 Ref2V reference-to-video finds relative scale between characters/objects hard to keep consistent — even within the same shot.
Prompting like '<subject 1> is slightly taller than <subject 2>' has limited success. They ask whether the model can read a scale bar or centimeter measurements from a character sheet, and note first/last frame references help but don't fit every project. Seeking community best practices.
More from Multimodal
- Midjourney dual --sref combo produces medieval woodcut cat knight art — sergeantsref · 2026-11-02
- Fluffy AI-generated character designs: fox girl, alien bug and cat photographer — OVolosin82152 · 2026-11-02
- Fable 5.5 reportedly one-shots animation in pure code as Anthropic's coding bet pays off — Hesamation · 2026-11-02
- FlowHMR gets 5x faster, demoing an LLM agent pipeline from video to photoreal mocap — camenduru · 2026-11-02
- Retro 1973 music video "Rock is Dead" by TECHHALLA showcases startlingly realistic AI generation — Mother-Pear-2392 · 2026-11-02
- Weekend splatting with COLMAP and Lichtfeld — janusch_patas · 2026-11-02