Spaces:
Running on Zero
Running on Zero
|
Download README.md from linoyts/video-replace-background: direct link, hf CLI and curl.
- Browser
- Download file 1.34 kB
-
https://huggingface.co/spaces/linoyts/video-replace-background/resolve/main/README.md
- Command line
-
hf download hf://spaces/linoyts/video-replace-background/README.md
-
curl -L -o README.md https://huggingface.co/spaces/linoyts/video-replace-background/resolve/main/README.md
1.34 kB
A newer version of the Gradio SDK is available: 6.31.0
metadata
title: Video Replace Background
emoji: 🎬
colorFrom: indigo
colorTo: green
sdk: gradio
sdk_version: 6.29.1
app_file: app.py
pinned: false
short_description: replace video background with LTX alpha LoRA + H3 inpainting
suggested_hardware: zero-a10g
Subject on a New Background
Put a video subject on a new, generated background, by chaining two models:
- Lift the subject with the LTX-2.5 Alpha-Gen matte LoRA (
hugging-apps/ltx-2-5-alpha-matte), called over the gradio API: video in, alpha matte + green-screen out. - Regenerate the background with MiniMax-H3 ref2va inpainting, held in this Space. The subject is handed to H3
as conditioning; the inverted matte marks the background as the region to repaint. Text encoding runs on the
qwen3vl-conditionerSpace, so this one never loads a 60 GiB encoder.
Upload a clip, describe the background, and press Replace background - or click one of the examples. Relight strength is the quality lever: at 0 the subject is held exactly (best when the new background's light matches the original); turn it up and H3 lights the subject into the new scene so it does not read as a cutout, while keeping identity.
Runs on ZeroGPU with the ref2v turbo LoRA at 8 steps. The matte and conditioner calls bill the caller's ZeroGPU quota, same as the denoise.