Optimization-based Neural Style Transfer - Revision history

Psuri8 at 22:28, 4 October 2018

2018-10-04T22:28:24Z

Psuri8 at 22:27, 4 October 2018

2018-10-04T22:27:54Z

Psuri8 at 21:25, 4 October 2018

2018-10-04T21:25:46Z

Psuri8: /* Usage */

2018-10-04T21:13:05Z

Usage

Psuri8 at 19:17, 24 July 2018

2018-07-24T19:17:36Z

Psuri8: /* (Optional) */

2018-07-24T19:14:54Z

(Optional)

Psuri8 at 19:12, 24 July 2018

2018-07-24T19:12:31Z

Psuri8 at 19:07, 24 July 2018

2018-07-24T19:07:49Z

Psuri8: Neural Style Transfer implementation in Python using Keras

2018-07-24T18:17:19Z

Neural Style Transfer implementation in Python using Keras

New page

'''OVERVIEW'''
This page provides instructions on how to set up and run an artistic style transfer program on your machines. The following implementation is based on the optimization method proposed by Gatys et al in 2015 ([https://arxiv.org/abs/1508.06576]).

'''SETUP'''
'''Pre-requisites'''
This program has been implemented in Python 2.7. The following libraries must be installed on your system to run it:

*Numpy
*Scipy
*Keras
*h5py
*Tensorflow

'''Getting Started'''
# Download the following zip file for the code {{File-DL|Test|zip}} to your preferred directory on your machine
# Open the Terminal from that directory and run the program using the following command: <blockquote> python NST.py path/to/content/image path/to/content/image path/to/style/image(s) path/to/generated/image </blockquote>
#; One or more style images can be provided by seperating each path with a space. Example: <blockquote> python NST.py images/inputs/content/Dipping-Sun.jpg images/inputs/style/the_scream.jpg images/inputs/style/wave_kanagawa.jpg Results/blue </blockquote>

@@ Line 14: / Line 14: @@
 *Tensorflow
-You can use the requirements.txt file provided in the zip folder with the command: <code> pip install requirements.txt </code> for easy installation.
+You can use the requirements.txt file provided in the zip folder with the command: <code> pip install requirements.txt</code>  for easy installation.
 This program can be run on the CPU or a GPU without any modifications. It was tested on MacOS and NVIDIA GTX 970 GPU running CUDA 9.0.

@@ Line 13: / Line 13: @@
 *h5py
 *Tensorflow
 This program can be run on the CPU or a GPU without any modifications. It was tested on MacOS and NVIDIA GTX 970 GPU running CUDA 9.0.
@@ Line 30: / Line 32: @@
 Open the Terminal from that directory and run the program using the following command:
-: <pre> python NST.py path/to/content/image path/to/content/image path/to/style/image(s) path/to/generated/image</pre>
+: <pre> python NST.py path/to/content-image path/to/style-image(s) path/to/generated-image</pre>
 Example:
 : <pre>python NST.py images/inputs/content/Dipping-Sun.jpg images/inputs/style/the_scream.jpg images/inputs/style/wave_kanagawa.jpg Results/blue</pre>

@@ Line 3: / Line 3: @@
 == Setup & Execution==
 ===Pre-requisites===
 This program has been implemented in Python 2.7. The following libraries must be installed on your system to run it:
@@ Line 11: / Line 13: @@
 *h5py
 *Tensorflow
 ===Usage===
@@ Line 47: / Line 51: @@
 : <pre>python NST.py images/inputs/content/Dipping-Sun.jpg images/inputs/style/the_scream.jpg images/inputs/style/wave_kanagawa.jpg Results/generated --num_iter 30 --style_weight 0.3 0.7 --content_weight 0.1</pre>
-===Results===
+==Results==
 The results are saved to the path provided at the beginning through the mandatory argument - Path to save generated image. The program saves the intermediate output after each iteration along with the final result (which is just the output after the final iteration).

@@ Line 13: / Line 13: @@
 ===Usage===
-Download the following zip file for the code {{Media:NeuralStyleTransfer.zip}} to your preferred directory on your machine. This program accepts multiple arguments that provide the user with the flexibility of altering inputs and outputs without modifying the code:
+Download the following zip file for the code {{Media:NeuralStyleTransfer.zip}} to your preferred directory on your machine.
 ====(Mandatory)====
 Without these arguments, the program will not run
@@ Line 20: / Line 22: @@
 *Path to style images(s). One or more style images can be provided by separating each path with space.
 *Path to save generated image.
 Open the Terminal from that directory and run the program using the following command:
@@ Line 26: / Line 30: @@
 : <pre>python NST.py images/inputs/content/Dipping-Sun.jpg images/inputs/style/the_scream.jpg images/inputs/style/wave_kanagawa.jpg Results/blue</pre>
 ====(Optional)====
 These arguments have default values, and can hence be skipped if required. However, to include it in your command, just append the syntax provided below to the main command:
@@ Line 41: / Line 46: @@
 For example, to run a multi-style transfer program with custom weights and 30 iterations, you can run the following command from your directory:
 : <pre>python NST.py images/inputs/content/Dipping-Sun.jpg images/inputs/style/the_scream.jpg images/inputs/style/wave_kanagawa.jpg Results/generated --num_iter 30 --style_weight 0.3 0.7 --content_weight 0.1</pre>

← Older revision		Revision as of 19:17, 24 July 2018
Line 38:		Line 38:
	*Model. Default value: "vgg19".		*Model. Default value: "vgg19".
	*: To change this, you can pass the following argument to the command: <br/> --model <new_value>. <br/> Example: <code>--model 'vgg16'</code>		*: To change this, you can pass the following argument to the command: <br/> --model <new_value>. <br/> Example: <code>--model 'vgg16'</code>
		+
		+	For example, to run a multi-style transfer program with custom weights and 30 iterations, you can run the following command from your directory:
		+	: <pre>python NST.py images/inputs/content/Dipping-Sun.jpg images/inputs/style/the_scream.jpg images/inputs/style/wave_kanagawa.jpg Results/generated --num_iter 30 --style_weight 0.3 0.7 --content_weight 0.1</pre>

@@ Line 1: / Line 1: @@
-'''OVERVIEW'''
+== Overview ==
 This page provides instructions on how to set up and run an artistic style transfer program on your machines. The following implementation is based on the optimization method proposed by Gatys et al in 2015 ([https://arxiv.org/abs/1508.06576]).
-'''SETUP'''
+== Setup & Execution==
-'''Pre-requisites'''
+===Pre-requisites===
 This program has been implemented in Python 2.7. The following libraries must be installed on your system to run it:
@@ Line 12: / Line 12: @@
 *Tensorflow
-'''Getting Started'''
+===Usage===
-# Download the following zip file for the code {{File-DL|Test|zip}} to your preferred directory on your machine
+Download the following zip file for the code {{File-DL|Test|zip}} to your preferred directory on your machine. This program accepts multiple arguments that provide the user with the flexibility of altering inputs and outputs without modifying the code:
-# Open the Terminal from that directory and run the program using the following command: <blockquote> python NST.py path/to/content/image path/to/content/image path/to/style/image(s) path/to/generated/image </blockquote>
-#; One or more style images can be provided by seperating each path with a space. Example: <blockquote> python NST.py images/inputs/content/Dipping-Sun.jpg images/inputs/style/the_scream.jpg images/inputs/style/wave_kanagawa.jpg Results/blue </blockquote>
+====(Mandatory)====