def compute_and_save_flow(video_data):
video_dir = "../data/videos/"
rgb_dir = "../data/rgb/"
flow_dir = "../data/flow/"
file_name = video_data["file_name"]
rgb_vid_in_p = str(video_dir + file_name + ".mp4")
rgb_4d_out_p = str(rgb_dir + file_name + ".npy")
flow_4d_out_p = str(flow_dir + file_name + ".npy")
if not is_file_here(rgb_vid_in_p):
return
if is_file_here(rgb_4d_out_p):
rgb_4d_out_p = None
if is_file_here(flow_4d_out_p):
flow_4d_out_p = None
if rgb_4d_out_p is None and flow_4d_out_p is None:
return
# Saves files to disk in format (time, height, width, channel) as numpy array
#flow_type = 1 # TVL1 optical flow
flow_type = None # will not process optical flow
op = OpticalFlow(rgb_vid_in_p=rgb_vid_in_p,
rgb_4d_out_p=rgb_4d_out_p,
flow_4d_out_p=flow_4d_out_p,
clip_flow_bound=20,
flow_type=flow_type)
op.process()
def video_to_numpy(url, vid_p, rgb_p, file_name):
rgb_vid_in_p = vid_p + file_name + ".mp4"
rgb_4d_out_p = rgb_p + file_name + ".npy"
op = OpticalFlow(rgb_vid_in_p=rgb_vid_in_p,
rgb_4d_out_p=rgb_4d_out_p,
flow_type=None)
op.process()
def main(argv):
if len(argv) < 2:
print("Usage: python show_video_transforms.py [video_file_path]")
return
# Read frames
op = OpticalFlow(rgb_vid_in_p=argv[1])
rgb_4d = op.vid_to_frames().astype(np.uint8) # ColorJitter need uint8
tl = TestLearner()
T = tl.get_transform("rgb", phase="train")
rgb_4d = T(rgb_4d).numpy().transpose(1, 2, 3, 0)
print(np.amin(rgb_4d), np.amax(rgb_4d))
print(rgb_4d.shape)
rgb_4d = cv2.normalize(rgb_4d, None, alpha=0, beta=255, norm_type=cv2.NORM_MINMAX, dtype=cv2.CV_32F)
op.frames_to_vid(rgb_4d, "../data/transformed.mp4")
import argparse
import cv2
import numpy as np
from optical_flow.optical_flow import OpticalFlow
of = OpticalFlow()
def main():
ap = argparse.ArgumentParser()
ap.add_argument('-i',
'--input',
default='videos\\cars.mp4',
help='Input video filename')
ap.add_argument(
'-m',
'--mode',
default='dense',
help=
'optical flow computation mode: sparse (Lucas Kanede), dense (Farneback)'
)
args = vars(ap.parse_args())
cap = cv2.VideoCapture(args['input'])
if not cap.isOpened():
print('[ERROR] Invalid video.')
return
# Take first frame and find features in it using ShiTomasi params for detection