async0x42 async0x42
async0x42
/ kimi-vl-a3b-instruct-video-inference-script-cli.py
Created
April 11, 2025 14:01
— forked from eek/kimi-vl-a3b-instruct-video-inference-script-cli.py
This script extracts frames from videos and generates descriptions using the Kimi-VL-A3B model. It takes the following arguments: video_path (required): Path to the input video file --max_frames (default=1): Maximum number of frames to extract --save_dir (default="./test-frames"): Directory to save extracted frames --prompt (default="Describe th…
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import cv2 | |
| import argparse | |
| import torch | |
| import os # Added import | |
| from PIL import Image | |
| from transformers import AutoModelForCausalLM, AutoProcessor | |
| # Function to extract frames from video, save them, and return paths | |
| def extract_frames(video_path, save_dir, target_fps=1, max_frames=1): | |
| """Extracts up to max_frames from a video file at target FPS, saves them, and returns their paths.""" |