else:
svm = classify_library.load_model('../data/models/svm_nopca.sav')
gmm_list = np.load(gmm_list + ".npz")['gmm_list']
index_class = np.load(class_index)['index_class']
index_class = index_class[()]
points = [] # a list of IDT features.
frame_lim = frame_step
for line in sys.stdin:
if line[0] != '[': # avoid getting info message as data
frame = int(line.split()[0])
if frame_lim <= frame:
frame_lim = frame_lim + frame_step
# print frame_lim<=frame
if points != []:
video_desc = IDT_feature.vid_descriptors(points)
fish = computeFV.create_fisher_vector_unsaved(
gmm_list, video_desc)
fish = np.array(fish).reshape(1, -1)
if args.no_pca:
result = svm.predict(fish)
else:
fish_pca = pca.transform(fish)
result = svm.predict(fish_pca)
print '\n' + 'RESULT: ' + OKGREEN + BOLD + index_class[
result[0]] + ENDC + '\n'
points = []
points.append(IDT_feature.IDTFeature(line))
video = args.video_name
vid = os.path.join(video_dir, video)
if not check_resolution(vid):
resizedName = os.path.join(tmp_dir, video)
if ffmpeg.resize(vid, resizedName):
vid = resizedName
command = dtBin + ' ' + vid
p = subprocess.Popen(command, shell=True, stdout=subprocess.PIPE)
raw_features, _ = p.communicate()
features = raw_features.split("\n")
points = []
for point in features:
if point != '' and point[
0] != '[': #to delete info messages from DenseTrackStab
points.append(IDT_feature.IDTFeature(point))
video_desc = IDT_feature.vid_descriptors(points)
gmm_list = np.load(gmm_list + ".npz")['gmm_list']
fish = computeFV.create_fisher_vector_unsaved(gmm_list, video_desc)
if not args.no_pca:
pca = classify_library.load_model('../data/models/pca.sav')
svm = classify_library.load_model('../data/models/svm.sav')
else:
svm = classify_library.load_model('../data/models/svm_nopca.sav')
fish = np.array(fish).reshape(1, -1)
if args.no_pca:
fish_pca = fish
else:
fish_pca = pca.transform(fish)
result = svm.predict(fish_pca)
