So this is what I have now:
As you can see, the neural style transfer thing is only going over the area the detection box is detecting. I am trying to put the transformed cool picture (which will always be less than 1200 x 900 because the detection box is 1200 x 900) in a black picture with dimensions 1200 x 900 so that I can save the video file.
My box is measured with: startX, endX, startY, and endY. The way I am trying to put the cool picture over the background right now is: black_background[startY:endY, startX:endX] = output, where output also has the size (endY - startY, endX - startX).
My way is not working, any insights? And also, for some reason, when I do "*black_background[startY:endY, startX:endX] = output", there is often a few pixel off broadcasting issue, like can't add (859, 100, 3) with (860, 100, 3). Is there a non-buggy solution to the black background issue? I feel like manually doing *black_background[startY:endY, startX:endX] = output is weird.
Here's my full code, I marked the if loop that actually matters with -----, thank you!
from __future__ import print_function
from imutils.video import VideoStream
from imutils.video import FPS
import numpy as np
import argparse
import imutils
import time
import cv2
from imutils import paths
import itertools
# We need to input model prototxt
ap = argparse.ArgumentParser()
ap.add_argument("-p", "--prototxt", required=True,
help="path to Caffe 'deploy' prototxt file")
ap.add_argument("-m", "--model", required=True,
help="path to Caffe pre-trained model")
ap.add_argument("-c", "--confidence", type=float, default=0.2,
help="minimum probability to filter weak detections")
ap.add_argument("-nm", "--neuralmodels", required=True,
help="path to directory containing neural style transfer models")
args = vars(ap.parse_args())
# we should identify the class first, and then transfer that block
CLASSES = ["background", "aeroplane", "bicycle", "bird", "boat",
"bottle", "bus", "car", "cat", "chair", "cow", "diningtable",
"dog", "horse", "motorbike", "person", "pottedplant", "sheep",
"sofa", "train", "tvmonitor"]
COLORS = np.random.uniform(0, 255, size=(len(CLASSES), 3))
# load our serialized model from disk
print("[INFO] loading model...")
DetectionNet = cv2.dnn.readNetFromCaffe(args["prototxt"], args["model"])
# grab the paths to all neural style transfer models in our 'models'
# directory, provided all models end with the '.t7' file extension
modelPaths = paths.list_files(args["neuralmodels"], validExts=(".t7",))
modelPaths = sorted(list(modelPaths))
# generate unique IDs for each of the model paths, then combine the
# two lists together
models = list(zip(range(0, len(modelPaths)), (modelPaths)))
# use the cycle function of itertools that can loop over all model
# paths, and then when the end is reached, restart again
modelIter = itertools.cycle(models)
(modelID, modelPath) = next(modelIter)
NTSnet = cv2.dnn.readNetFromTorch(modelPath)
# initialize the video stream, allow the cammera sensor to warmup,
# and initialize the FPS counter
print("[INFO] starting video stream...")
vs = VideoStream(src=1).start()
fps = FPS().start()
fourcc = cv2.VideoWriter_fourcc(*'XVID')
output_video = cv2.VideoWriter('output.avi', fourcc, 20.0, (1200, 900))
while True:
# grab the frame from the threaded video stream and resize it
# to have a maximum width of 400 pixels
frame = vs.read()
frame = imutils.resize(frame, width=1200, height=900)
# grab the frame dimensions and convert it to a blob
(h, w) = frame.shape[:2]
blob = cv2.dnn.blobFromImage(cv2.resize(frame, (300, 300)),
0.007843, (300, 300), 127.5)
# pass the blob through the network and obtain the detections and
# predictions
DetectionNet.setInput(blob)
detections = DetectionNet.forward()
# loop over the detections
for i in np.arange(0, detections.shape[2]):
# extract the confidence (i.e., probability) associated with
# the prediction
confidence = detections[0, 0, i, 2]
# filter out weak detections by ensuring the `confidence` is
# greater than the minimum confidence
if confidence > args["confidence"]:
# extract the index of the class label from the
# `detections`, then compute the (x, y)-coordinates of
# the bounding box for the object
idx = int(detections[0, 0, i, 1])
if(CLASSES[idx] == "person" and confidence > .90):
box = detections[0, 0, i, 3:7] * np.array([w, h, w, h])
(startX, startY, endX, endY) = box.astype("int")
# draw the prediction on the frame
label = "{}: {:.2f}%".format("PERSON",
confidence * 100)
cv2.rectangle(frame, (startX, startY), (endX, endY),
COLORS[idx], 2)
y = startY - 15 if startY - 15 > 15 else startY + 15
cv2.putText(frame, label, (startX, y),
cv2.FONT_HERSHEY_SIMPLEX, 0.5, COLORS[idx], 2)
# print box area in background
newimage = frame[startY:endY, startX:endX]
(h, w) = newimage.shape[:2]
#print(h,w)
#print(startX, endX, startY, endY)
noise_picture = cv2.imread('white_noise.jpg')
black_background = cv2.imread('black.png')
-------------------------------------------------------------------
*if(h > 0 and w > 0):
# to_be_transformed is the detection box area
# resize that area for MobileNetSSD
#to_be_transformed = imutils.resize(to_be_transformed, height=450)
(height_orig, width_orig) = noise_picture.shape[:2]
noise_picture[startY:endY, startX:endX] = newimage
noise_picture = imutils.resize(noise_picture, height=450)
# run it through the network, output is the image
(h, w) = noise_picture.shape[:2]
# print(h, w)
blob2 = cv2.dnn.blobFromImage(noise_picture, 1.0, (w, h), (103.939, 116.779, 123.680), swapRB=False, crop=False)
NTSnet.setInput(blob2)
output = NTSnet.forward()
output = output.reshape((3, output.shape[2], output.shape[3]))
output[0] += 103.939
output[1] += 116.779
output[2] += 123.680
output /= 255.0
output = output.transpose(1, 2, 0)
# set the 600 x 450 back to the original size
black_background = imutils.resize(black_background, width=1200, height = 900)
output = imutils.resize(output, width=1200)
#black_background[startY:endY, startX:endX] = output[startY:endY, startX:endX]
output = output[startY:endY, startX:endX]
(h2, w2) = output.shape[:2]
if(h2>0 and w2>0 ):
cv2.imshow('hmm', output)
black_background[startY:endY, startX:endX] = output
cv2.imshow("uh", black_background)
#output_video.write(black_background)
#output_video.write(background)*
---------------------------------------------------------------
# show the output frame, which is the whole thing
cv2.imshow("Frame", frame)
key = cv2.waitKey(1) & 0xFF
# if the `q` key was pressed, break from the loop
if key == ord("q"):
break
# update the FPS counter
fps.update()
# stop the timer and display FPS information
fps.stop()
print("[INFO] elapsed time: {:.2f}".format(fps.elapsed()))
print("[INFO] approx. FPS: {:.2f}".format(fps.fps()))
# do a bit of cleanup
cv2.destroyAllWindows()
vs.stop()
