Spaces:
Sleeping
Sleeping
File size: 2,408 Bytes
3a76a8c 2899e27 3a76a8c 621c3a4 3a76a8c 915ddca 3a76a8c 915ddca 3a76a8c 915ddca 3a76a8c 2e77929 6733e56 d0d4c48 91671ee 3a76a8c aa4e1bc 3a76a8c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 |
import os
import tensorflow as tf
os.environ['TFHUB_MODEL_LOAD_FORMAT'] = 'COMPRESSED'
import numpy as np
import PIL.Image
import gradio as gr
import tensorflow_hub as hub
import matplotlib.pyplot as plt
hub_module = hub.load('https://tfhub.dev/google/magenta/arbitrary-image-stylization-v1-256/2')
def tensor_to_image(tensor):
tensor = tensor*255
tensor = np.array(tensor, dtype=np.uint8)
if np.ndim(tensor) > 3:
assert tensor.shape[0] == 1
tensor = tensor[0]
return PIL.Image.fromarray(tensor)
content_image_input = gr.Image(label="Content Image", sources = ('upload', 'webcam'))
style_image_input = gr.Image(label="Style Image")
text = "#Developer: Aditya Jadhav #College: Government Polytechnic Nagpur #Branch: AIML (Artificial Intelligence and Machine Learning), 3rd year #Date: 17th November 2023 #LinkedIn: https://www.linkedin.com/in/aditya-jadhav-27a790269 #GitHub: https://github.com/AdityaJ9801 #Introduction: The application focuses on neural style transfer, where the style from a style image is applied to a content image."
def perform_neural_transfer(content_image_input, style_image_input):
if content_image_input is None:
# Handle the case when content image is not provided
return PIL.Image.fromarray(np.zeros((1, 1, 3), dtype=np.uint8))
# Load content images
content_image = content_image_input.astype(np.float32)[np.newaxis, ...] / 255.
if style_image_input is None:
# Handle the case when style image is not provided
return tensor_to_image(content_image)
style_image = style_image_input.astype(np.float32)[np.newaxis, ...] / 255.
# Apply neural style transfer
outputs = hub_module(tf.constant(content_image), tf.constant(style_image))
stylized_image = outputs[0]
return tensor_to_image(stylized_image)
app_interface = gr.Interface(
fn=perform_neural_transfer,
inputs=[content_image_input, style_image_input],
outputs="image",
title="Art Generation with Neural Style Transfer",
description=text,
concurrency_limit = 5,
theme = gr.themes.Default(),
examples=[
['content1.jpg','image1.jpg'],
[None,'image2.jpg'],
[None,'image4.jpg'],
[None,'image5.jpg'],
[None,'image6.jpg'],
[None,'image7.jpg'],
[None,'image8.jpg'],
]
)
app_interface.launch(debug =True)
|