zhang-zx/AVID
This respository contains the code for the CVPR 2024 paper AVID: Any-Length Video Inpainting with Diffusion Model.
This tool helps video editors and content creators seamlessly remove unwanted objects or fill in missing areas within video footage. You provide a video, specify the region to change in the first frame, and describe the desired outcome with a text prompt. The output is an edited video where the specified area is intelligently inpainted, maintaining visual and temporal consistency across the entire clip.
177 stars. No commits in the last 6 months.
Use this if you need to erase objects, replace backgrounds, or repair damaged sections in videos of any length using a text description, and require the edits to flow smoothly over time.
Not ideal if you need to perform precise, frame-by-frame manual touch-ups or require extremely fine-grained control over individual pixels without text guidance.
Stars
177
Forks
—
Language
—
License
MIT
Category
Last pushed
Feb 27, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/zhang-zx/AVID"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
hao-ai-lab/FastVideo
A unified inference and post-training framework for accelerated video generation.
ModelTC/LightX2V
Light Image Video Generation Inference Framework
thu-ml/TurboDiffusion
TurboDiffusion: 100–200× Acceleration for Video Diffusion Models
PKU-YuanGroup/Helios
Helios: Real Real-Time Long Video Generation Model
PKU-YuanGroup/MagicTime
[TPAMI 2025🔥] MagicTime: Time-lapse Video Generation Models as Metamorphic Simulators