0

I have posted code on this site before and I learnt that I can't post the whole thing. So, I will only post the code that matters.

So, what I am trying to do is to take an object detector(for images) and applying it to each frame of a given video.

The only thing is that I don't know how to finish it up. That is, once I detect the first frame what do I do with this frame? Do I store it somewhere? What do I do with the other frames? And once I handle these frames how do I recombine these frames into a video ie the output video?

Here is the code:

import numpy as np
import cv2
from numpy import expand_dims
from keras.models import load_model
from keras.preprocessing.image import load_img
from keras.preprocessing.image import img_to_array
from matplotlib import pyplot
from matplotlib.patches import Rectangle

model = load_model('model.h5')

# define the expected input shape for the model
input_w, input_h = 416, 416

# define the anchors
anchors = [[116,90, 156,198, 373,326], [30,61, 62,45, 59,119], [10,13, 16,30, 33,23]]

# define the labels
labels = ["person", "bicycle", "car", "motorbike", "aeroplane", "bus", "train", "truck",
    "boat", "traffic light", "fire hydrant", "stop sign", "parking meter", "bench",
    "bird", "cat", "dog", "horse", "sheep", "cow", "elephant", "bear", "zebra", "giraffe",
    "backpack", "umbrella", "handbag", "tie", "suitcase", "frisbee", "skis", "snowboard",
    "sports ball", "kite", "baseball bat", "baseball glove", "skateboard", "surfboard",
    "tennis racket", "bottle", "wine glass", "cup", "fork", "knife", "spoon", "bowl", "banana",
    "apple", "sandwich", "orange", "broccoli", "carrot", "hot dog", "pizza", "donut", "cake",
    "chair", "sofa", "pottedplant", "bed", "diningtable", "toilet", "tvmonitor", "laptop", "mouse",
    "remote", "keyboard", "cell phone", "microwave", "oven", "toaster", "sink", "refrigerator",
    "book", "clock", "vase", "scissors", "teddy bear", "hair drier", "toothbrush"]

vs = cv2.VideoCapture('video.mp4')

class_threshold = 0.6
boxes = list()

while True:
    (grabbed, frame) = vs.read()

    if not grabbed:
        break

    if W is None or H is None:
        (H, W) = frame.shape[:2]

    image, image_w, image_h = load_image_pixels(frame, (input_w, input_h))
    yhat = model.predict(image)

    for i in range(len(yhat)):
        # decode the output of the network
        boxes += decode_netout(yhat[i][0], anchors[i], class_threshhold, input_h, input_w)
    # correct the sizes of the bounding boxes for the shape of the image
    correct_yolo_boxes(boxes, image_h, image_w, input_h, input_w)
    # suppress non-maximal boxes
    do_nms(boxes, 0.5)

    # get the details of the detected objects
    v_boxes, v_labels, v_scores = get_boxes(boxes, labels, class_threshold)

    # draw what we found
    draw_boxes(frame, v_boxes, v_labels, v_scores)
ysquared
  • 125
  • 1
  • 2
  • 8

1 Answers1

1

You can use the VideoWriter from opencv to output the frames again as a video.

Some example code on how to use it:

fourcc = cv2.VideoWriter_fourcc(*'XVID')
video_writer = cv2.VideoWriter('test.avi', fourcc, 30, (image_w, image_h))
...
while True:
    ....
    video_writer.write(frame)
    ....
....
video_writer.release()

For reference openCV video saving in python

Nopileos
  • 1,976
  • 7
  • 17
  • `draw_boxes(frame, v_boxes, v_labels, v_scores)` (look at my code please) is responsible to detect the object from each frame. do I have to assign this to a variable? or what do I do with this function call as the while loop is active ie looping over each frame? how do I then tie this with videowriter? – ysquared Jan 09 '20 at 16:58
  • @Job I assumed that `draw_boxes` will draw the boxes into the frame. So after you called draw_boxes the frame contains them. If that is the case you just have to add `video_writer.write(frame)` after it. – Nopileos Jan 09 '20 at 21:26