elif dataset.upper().startswith('KTH'):
from lib.dataloaders.kth_dataset import Video_Dataset_small_clip
else:
raise ('Unknown Dataset')
dataset_frames = os.path.abspath(
os.path.join(root_path, dataset_cfg.dataset.dataset_frames_folder))
boxes_file = os.path.abspath(
os.path.join(root_path, dataset_cfg.dataset.boxes_file))
split_txt_path = os.path.abspath(
os.path.join(root_path, dataset_cfg.dataset.split_txt_path))
### get videos id
actions = dataset_cfg.dataset.classes
cls2idx = {actions[i]: i for i in range(0, len(actions))}
vid2idx, vid_names = get_vid_dict(dataset_frames)
# # get mean
# mean = [112.07945832, 112.87372333, 106.90993363] # ucf-101 24 classes
mean = [0.5, 0.5, 0.5]
std = [0.5, 0.5, 0.5]
spatial_transform = Compose([
Scale(sample_size), # [Resize(sample_size),
ToTensor(),
Normalize(mean, std)
])
temporal_transform = LoopPadding(sample_duration)
n_classes = len(actions)
mean = [112.07945832, 112.87372333, 106.90993363] # ucf-101 24 classes
# generate model
actions = [
'__background__', 'Basketball', 'BasketballDunk', 'Biking',
'CliffDiving', 'CricketBowling', 'Diving', 'Fencing',
'FloorGymnastics', 'GolfSwing', 'HorseRiding', 'IceDancing',
'LongJump', 'PoleVault', 'RopeClimbing', 'SalsaSpin', 'SkateBoarding',
'Skiing', 'Skijet', 'SoccerJuggling', 'Surfing', 'TennisSwing',
'TrampolineJumping', 'VolleyballSpiking', 'WalkingWithDog'
]
cls2idx = {actions[i]: i for i in range(0, len(actions))}
### get videos id
vid2idx, vid_names = get_vid_dict(dataset_folder)
spatial_transform = Compose([
Scale(sample_size), # [Resize(sample_size),
ToTensor(),
Normalize(mean, [1, 1, 1])
])
temporal_transform = LoopPadding(sample_duration)
n_classes = len(actions)
##########################################
# Model Initialization #
##########################################
model = Model(actions, sample_duration, sample_size)