|
import gradio as gr |
|
import numpy as np |
|
import imageio |
|
|
|
import tensorflow as tf |
|
from tensorflow import keras |
|
|
|
from utils import TubeMaskingGenerator |
|
from utils import read_video, frame_sampling, denormalize, reconstrunction |
|
from utils import IMAGENET_MEAN, IMAGENET_STD, num_frames, patch_size, input_size |
|
from labels import K400_label_map, SSv2_label_map, UCF_label_map |
|
|
|
|
|
LABEL_MAPS = { |
|
'K400': K400_label_map, |
|
'SSv2': SSv2_label_map, |
|
'UCF' : UCF_label_map |
|
} |
|
|
|
ALL_MODELS = [ |
|
'TFVideoMAE_L_K400_16x224', |
|
'TFVideoMAE_B_SSv2_16x224', |
|
'TFVideoMAE_B_UCF_16x224', |
|
] |
|
|
|
sample_example = [ |
|
["examples/k400.mp4", ALL_MODELS[0], 0.9], |
|
["examples/ssv2.avi", ALL_MODELS[1], 0.8], |
|
["examples/ucf.mp4", ALL_MODELS[2], 0.7], |
|
] |
|
|
|
def tube_mask_generator(mask_ratio): |
|
window_size = ( |
|
num_frames // 2, |
|
input_size // patch_size[0], |
|
input_size // patch_size[1] |
|
) |
|
tube_mask = TubeMaskingGenerator( |
|
input_size=window_size, |
|
mask_ratio=mask_ratio |
|
) |
|
make_bool = tube_mask() |
|
bool_masked_pos_tf = tf.constant(make_bool, dtype=tf.int32) |
|
bool_masked_pos_tf = tf.expand_dims(bool_masked_pos_tf, axis=0) |
|
bool_masked_pos_tf = tf.cast(bool_masked_pos_tf, tf.bool) |
|
return bool_masked_pos_tf |
|
|
|
|
|
def get_model(model_type): |
|
|
|
|
|
|
|
|
|
|
|
|
|
ft_model = keras.models.load_model(model_type + '_FT') |
|
pt_model = keras.models.load_model(model_type + '_PT') |
|
|
|
if 'K400' in model_type: |
|
data_type = 'K400' |
|
elif 'SSv2' in model_type: |
|
data_type = 'SSv2' |
|
else: |
|
data_type = 'UCF' |
|
|
|
label_map = LABEL_MAPS.get(data_type) |
|
label_map = {v: k for k, v in label_map.items()} |
|
|
|
return ft_model, pt_model, label_map |
|
|
|
|
|
def inference(video_file, model_type, mask_ratio): |
|
|
|
container = read_video(video_file) |
|
frames = frame_sampling(container, num_frames=num_frames) |
|
|
|
|
|
bool_masked_pos_tf = tube_mask_generator(mask_ratio) |
|
ft_model, pt_model, label_map = get_model(model_type) |
|
ft_model.trainable = False |
|
pt_model.trainable = False |
|
|
|
|
|
outputs_ft = ft_model(frames[None, ...], training=False) |
|
probabilities = tf.nn.softmax(outputs_ft).numpy().squeeze(0) |
|
confidences = { |
|
label_map[i]: float(probabilities[i]) for i in np.argsort(probabilities)[::-1] |
|
} |
|
|
|
|
|
outputs_pt = pt_model(frames[None, ...], bool_masked_pos_tf, training=False) |
|
reconstruct_output, mask = reconstrunction( |
|
frames[None, ...], bool_masked_pos_tf, outputs_pt |
|
) |
|
|
|
|
|
input_frame = denormalize(frames) |
|
input_mask = denormalize(mask[0] * frames) |
|
output_frame = denormalize(reconstruct_output) |
|
|
|
frames = [] |
|
for frame_a, frame_b, frame_c in zip(input_frame, input_mask, output_frame): |
|
combined_frame = np.hstack([frame_a, frame_b, frame_c]) |
|
frames.append(combined_frame) |
|
|
|
combined_gif = 'combined.gif' |
|
imageio.mimsave(combined_gif, frames, duration=300, loop=0) |
|
return confidences, combined_gif |
|
|
|
|
|
def main(): |
|
iface = gr.Interface( |
|
fn=inference, |
|
inputs=[ |
|
gr.Video(type="file", label="Input Video"), |
|
gr.Dropdown( |
|
choices=ALL_MODELS, |
|
default="TFVideoMAE_L_K400_16x224", |
|
label="Model" |
|
), |
|
gr.Slider( |
|
0.5, |
|
1.0, |
|
step=0.1, |
|
default=0.5, |
|
label='Mask Ratio' |
|
) |
|
], |
|
outputs=[ |
|
gr.Label(num_top_classes=3, label='scores'), |
|
gr.Image(type="filepath", label='reconstructed') |
|
], |
|
examples=sample_example, |
|
title="VideoMAE", |
|
description="Keras reimplementation of <a href='https://github.com/innat/VideoMAE'>VideoMAE</a> is presented here." |
|
) |
|
|
|
iface.launch() |
|
|
|
if __name__ == '__main__': |
|
main() |