1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2025-02-09 14:14:39 +02:00
FFmpeg/libavfilter
Ting Fu 13f5613e68 dnn_backend_native_layer_mathunary: add atan support
It can be tested with the model generated with below python script:

import tensorflow as tf
import numpy as np
import imageio

in_img = imageio.imread('input.jpeg')
in_img = in_img.astype(np.float32)/255.0
in_data = in_img[np.newaxis, :]

x = tf.placeholder(tf.float32, shape=[1, None, None, 3], name='dnn_in')
x1 = tf.atan(x)
x2 = tf.divide(x1, 3.1416/4) # pi/4
y = tf.identity(x2, name='dnn_out')

sess=tf.Session()
sess.run(tf.global_variables_initializer())

graph_def = tf.graph_util.convert_variables_to_constants(sess, sess.graph_def, ['dnn_out'])
tf.train.write_graph(graph_def, '.', 'image_process.pb', as_text=False)

print("image_process.pb generated, please use \
path_to_ffmpeg/tools/python/convert.py to generate image_process.model\n")

output = sess.run(y, feed_dict={x: in_data})
imageio.imsave("out.jpg", np.squeeze(output))

Signed-off-by: Ting Fu <ting.fu@intel.com>
Signed-off-by: Guo Yejun <yejun.guo@intel.com>
2020-06-25 08:41:50 +08:00
..
2020-02-14 09:59:27 +01:00
2020-05-23 15:50:20 +02:00
2020-05-30 19:02:34 +08:00
2020-04-30 12:18:36 +02:00
2020-03-17 22:46:36 +01:00
2020-05-30 18:04:14 +02:00
2020-02-07 17:07:30 +01:00
2020-03-17 22:46:36 +01:00
2020-03-17 22:46:36 +01:00
2020-05-23 15:50:20 +02:00
2020-05-30 18:04:14 +02:00
2020-02-14 21:49:47 +01:00
2020-03-17 22:46:36 +01:00
2020-02-04 18:28:04 +01:00
2020-02-29 22:31:01 +01:00
2020-05-30 18:04:14 +02:00
2020-03-17 22:46:36 +01:00
2020-03-17 22:46:36 +01:00
2020-03-17 22:46:36 +01:00
2020-03-17 22:46:36 +01:00
2020-03-17 22:46:36 +01:00
2020-03-17 22:46:36 +01:00
2020-03-17 22:46:36 +01:00
2020-02-04 18:28:04 +01:00
2020-05-23 15:52:27 +02:00
2020-03-17 22:46:36 +01:00
2020-04-18 12:34:49 +02:00
2020-03-17 22:46:36 +01:00
2020-03-17 22:46:36 +01:00