Person Remover LoRA for MiniMax H3 Released with Full ComfyUI Workflow
linoy_tsaban · x · 2026-10-08
Akatz Labs released Person Remover LoRA V1 built on MiniMax H3 (Ref2VA): select a person in a video, track and mask them with SAM 3.1, fill the mask green, and let the model generate the replacement background.
Workflow essentials: use a ComfyUI with native MiniMax H3 and SAM 3.1 support, update H3 Relay for window preview and reroll controls; place the LoRA in ComfyUI/models/loras/; drag in the example JSON workflow; input a 24fps video with width/height divisible by 32, starting with a 5-second continuous shot; supply a clean first frame with the person removed (made in an image editor or via an image-editing model). The author is also using it to generate training samples for a Character Swap V2 dataset, with an updated workflow for videos longer than 5s coming tomorrow.
More from Multimodal
- D-OPCD distills agent harness gains into diffusion weights, lifting T2I scores 60.52 to 65.09 — Wenxuan Wang · 2026-10-08
- QuadTok: quadtree visual tokenizer saves 10% tokens, hits 2.08 gFID on ImageNet 256 — Yucheng Mao · 2026-10-08
- Seedance 2.5 Airport Makeup Comedy Video and Its Full 30-Second Prompt, Shared — SimplyAnnisa · 2026-10-08
- Designer shares Midjourney v8.2 style prompt and generated results — azed_ai · 2026-10-08
- Kling 4.0 Flash demo: one-prompt Christmas video shared by AI creator — umesh_ai · 2026-10-08
- LTX Alpha Matte LoRA + H3 Inpainting enable surprisingly good video background replacement — linoy_tsaban · 2026-10-08