Spaces:
Sleeping
Sleeping
File size: 1,953 Bytes
85b6267 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 |
import cv2
import numpy as np
from utils.imageEmbedding.index import get_image_embedding
from utils.imageToText.index import extract_text
import requests
def get_video_embedding(video_url):
try:
cap = cv2.VideoCapture(video_url)
fps = cap.get(cv2.CAP_PROP_FPS)
interval = int(fps) # Capture a frame every second
frame_count = 0
video_embeddings = []
while(cap.isOpened()):
ret, frame = cap.read()
if ret:
if frame_count % interval == 0:
# Convert frame to binary format
ret, buffer = cv2.imencode('.jpg', frame)
if not ret:
continue
# Convert frame binary data to bytes
frame_bytes = buffer.tobytes()
# Call the route to get image embedding and extracted text
response = requests.post('https://imageprocessing-backend.hf.space/extract_image_text_and_embedding_binary_data', data=frame_bytes)
if response.status_code != 200:
print(f"Failed to process image: {frame_bytes}")
continue
result = response.json()
image_embedding = result.get("image_embedding")
extracted_text = result.get("extracted_text")
video_embeddings.append({"image_embedding": image_embedding, "extracted_text": extracted_text})
frame_count += 1
else:
break
cap.release()
return video_embeddings
except Exception as e:
print(e)
# Example usage:
# video_url = "https://utfs.io/f/ef6c037f-fa61-471a-8956-562bc2d62531-fzxs1i.mp4"
# video_embeddings = get_video_embedding(video_url)
# print("Video Embeddings:", video_embeddings) |