Two-node package wrapping the STAR (ICCV 2025) diffusion-based video upscaling pipeline: - STAR Model Loader: loads UNet+ControlNet, OpenCLIP text encoder, and temporal VAE with auto-download from HuggingFace - STAR Video Super-Resolution: runs the full diffusion pipeline with configurable upscale factor, guidance, solver mode, chunking, and color correction Includes three VRAM offload modes (disabled/model/aggressive) to support GPUs from 12GB to 40GB+. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
11 lines
415 B
Python
11 lines
415 B
Python
import os
|
|
import subprocess
|
|
import sys
|
|
|
|
req_file = os.path.join(os.path.dirname(os.path.realpath(__file__)), "requirements.txt")
|
|
subprocess.check_call([sys.executable, "-m", "pip", "install", "-r", req_file])
|
|
|
|
star_dir = os.path.join(os.path.dirname(os.path.realpath(__file__)), "STAR")
|
|
if not os.path.isdir(star_dir):
|
|
subprocess.check_call(["git", "clone", "https://github.com/NJU-PCALab/STAR.git", star_dir])
|