Duplicate from nightfury/Colorizer_Models

Co-authored-by: spartan <nightfury@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,31 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,14 @@
+---
+title: Colorizer Models
+emoji: 🌈🎨
+colorFrom: red
+colorTo: orange
+sdk: gradio
+sdk_version: 3.5
+app_file: app.py
+pinned: false
+license: bsd-2-clause
+duplicated_from: nightfury/Colorizer_Models
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,143 @@
+import gradio as gr
+import numpy as np
+import colorizers as c
+
+from colorizers.util import postprocess_tens, preprocess_img
+
+def interface(image, model: str = "eccv16"):
+    if model == "eccv16":
+        img = c.eccv16(pretrained=True).eval()
+    else:
+        img = c.siggraph17(pretrained=True).eval()
+    oimg = np.asarray(image)
+    if(oimg.ndim == 2):
+        oimg = np.tile(oimg[:,:,None], 3)
+    (tens_l_orig, tens_l_rs) = preprocess_img(oimg)
+
+    output_img = postprocess_tens(
+        tens_l_orig,
+        img(tens_l_rs).cpu()
+    )
+    return output_img
+
+css='''
+.Box {
+	background-color: var(--color-canvas-default);
+	border-color: var(--color-border-default);
+	border-style: solid;
+	border-width: 1px;
+	border-radius: 6px;
+}
+.d-flex {
+	display: flex !important;
+}
+.flex-md-row {
+	flex-direction: row !important;
+}
+.flex-column {
+	flex-direction: column !important;
+}
+'''
+title = "Image Colorization Using AI Models"
+description = r"""<center>An automatic colorization functionality for Real-Time User-Guided Image Colorization with Learned Deep Priors,ECCV16 & SIGGRAPH 2017 Models!<br>
+Practically the algorithm is used to COLORIZE your **old BLACK & WHITE / GRAYSCALE photos**.<br>
+To use it, simply just upload the concerned image.<br>
+"""
+article = r"""
+<p style='text-align: '> Given a grayscale photograph as input, this demo attacks the problem of hallucinating a plausible color version of the photograph. This problem is clearly underconstrained, so previous approaches have either relied on significant user interaction or resulted in desaturated colorizations. A fully automatic approach has been proposed that produces vibrant and realistic colorizations. The underlying uncertainty of the problem was embraced by posing it as a classification task and use class-rebalancing at training time to increase the diversity of colors in the result. The system is implemented as a feed-forward pass in a CNN at test time and is trained on over a million color images. The algorithm is evaluated using a "colorization Turing test," asking human participants to choose between a generated and ground truth color image. The method used here successfully fools humans on 32% of the trials, significantly higher than other methodology used by the other photo automation tools. Moreover, the colorization can be a powerful pretext task for self-supervised feature learning, acting as a cross-channel encoder. This approach results in state-of-the-art performance on several feature learning benchmarks. </p>
+<img src="https://camo.githubusercontent.com/8f849fd53753e61659664d37849703d949209ec93957d1a2fd529be9772f7030/687474703a2f2f726963687a68616e672e6769746875622e696f2f636f6c6f72697a6174696f6e2f7265736f75726365732f696d616765732f746561736572342e6a7067" alt="Teaser Image" data-canonical-src="http://richzhang.github.io/colorization/resources/images/teaser4.jpg" style="max-width: 100%;">
+<p>
+  <img class="round" style="height:275px" src="http://richzhang.github.io/colorization/resources/images/net_diagram.jpg">
+</p>
+<div class="footer">
+  <p>Built by <a href="" style="text-decoration: underline;" target="_blank">Dushyantsinh Jadeja</a> & the Model being used in here is from <a href="https://github.com/richzhang/colorization" style="text-decoration: underline;" target="_blank">richzhang</a> - Gradio Demo by 🤗 Hugging Face </p>
+</div>
+<div class="acknowledgments">
+  <p>
+  <h4>LICENSE</h4>
+  </p>
+  <h3 class="mt-0 mb-2 h4">BSD 2-Clause "Simplified" License</h3>
+  <h4 class="mt-1 mb-2 h5">Permissions</h4>
+  <ul class="list-style-none">
+    <li class="text-small pl-3">
+      <svg width="13" class="octicon octicon-check color-fg-success ml-n3 v-align-middle" viewBox="0 0 16 16" version="1.1" height="13" aria-hidden="true">
+        <path fill-rule="evenodd" d="M13.78 4.22a.75.75 0 010 1.06l-7.25 7.25a.75.75 0 01-1.06 0L2.22 9.28a.75.75 0 011.06-1.06L6 10.94l6.72-6.72a.75.75 0 011.06 0z"></path>
+      </svg>
+      <span class="v-align-middle" title="The licensed material and derivatives may be used for commercial purposes."> Commercial use </span>
+    </li>
+    <li class="text-small pl-3">
+      <svg width="13" class="octicon octicon-check color-fg-success ml-n3 v-align-middle" viewBox="0 0 16 16" version="1.1" height="13" aria-hidden="true">
+        <path fill-rule="evenodd" d="M13.78 4.22a.75.75 0 010 1.06l-7.25 7.25a.75.75 0 01-1.06 0L2.22 9.28a.75.75 0 011.06-1.06L6 10.94l6.72-6.72a.75.75 0 011.06 0z"></path>
+      </svg>
+      <span class="v-align-middle" title="The licensed material may be modified."> Modification </span>
+    </li>
+    <li class="text-small pl-3">
+      <svg width="13" class="octicon octicon-check color-fg-success ml-n3 v-align-middle" viewBox="0 0 16 16" version="1.1" height="13" aria-hidden="true">
+        <path fill-rule="evenodd" d="M13.78 4.22a.75.75 0 010 1.06l-7.25 7.25a.75.75 0 01-1.06 0L2.22 9.28a.75.75 0 011.06-1.06L6 10.94l6.72-6.72a.75.75 0 011.06 0z"></path>
+      </svg>
+      <span class="v-align-middle" title="The licensed material may be distributed."> Distribution </span>
+    </li>
+    <li class="text-small pl-3">
+      <svg width="13" class="octicon octicon-check color-fg-success ml-n3 v-align-middle" viewBox="0 0 16 16" version="1.1" height="13" aria-hidden="true">
+        <path fill-rule="evenodd" d="M13.78 4.22a.75.75 0 010 1.06l-7.25 7.25a.75.75 0 01-1.06 0L2.22 9.28a.75.75 0 011.06-1.06L6 10.94l6.72-6.72a.75.75 0 011.06 0z"></path>
+      </svg>
+      <span class="v-align-middle" title="The licensed material may be used and modified in private."> Private use </span>
+    </li>
+  </ul>
+  <h4 class="mt-1 mb-2 h5">Limitations</h4>
+  <ul class="list-style-none">
+    <li class="text-small pl-3">
+      <svg width="13" class="octicon octicon-x color-fg-danger ml-n3 v-align-middle" viewBox="0 0 16 16" version="1.1" height="13" aria-hidden="true">
+        <path fill-rule="evenodd" d="M3.72 3.72a.75.75 0 011.06 0L8 6.94l3.22-3.22a.75.75 0 111.06 1.06L9.06 8l3.22 3.22a.75.75 0 11-1.06 1.06L8 9.06l-3.22 3.22a.75.75 0 01-1.06-1.06L6.94 8 3.72 4.78a.75.75 0 010-1.06z"></path>
+      </svg>
+      <span class="v-align-middle" title="This license includes a limitation of liability."> Liability </span>
+    </li>
+    <li class="text-small pl-3">
+      <svg width="13" class="octicon octicon-x color-fg-danger ml-n3 v-align-middle" viewBox="0 0 16 16" version="1.1" height="13" aria-hidden="true">
+        <path fill-rule="evenodd" d="M3.72 3.72a.75.75 0 011.06 0L8 6.94l3.22-3.22a.75.75 0 111.06 1.06L9.06 8l3.22 3.22a.75.75 0 11-1.06 1.06L8 9.06l-3.22 3.22a.75.75 0 01-1.06-1.06L6.94 8 3.72 4.78a.75.75 0 010-1.06z"></path>
+      </svg>
+      <span class="v-align-middle" title="This license explicitly states that it does NOT provide any warranty."> Warranty </span>
+    </li>
+  </ul>
+  <h4 class="mt-1 mb-2 h5">Conditions</h4>
+  <ul class="list-style-none">
+    <li class="text-small pl-3">
+      <svg width="13" class="octicon octicon-info color-fg-accent ml-n3 v-align-middle" viewBox="0 0 16 16" version="1.1" height="13" aria-hidden="true">
+        <path fill-rule="evenodd" d="M8 1.5a6.5 6.5 0 100 13 6.5 6.5 0 000-13zM0 8a8 8 0 1116 0A8 8 0 010 8zm6.5-.25A.75.75 0 017.25 7h1a.75.75 0 01.75.75v2.75h.25a.75.75 0 010 1.5h-2a.75.75 0 010-1.5h.25v-2h-.25a.75.75 0 01-.75-.75zM8 6a1 1 0 100-2 1 1 0 000 2z"></path>
+      </svg>
+      <span class="v-align-middle" title="A copy of the license and copyright notice must be included with the licensed material."> License and copyright notice </span>
+    </li>
+  </ul>
+</div> For the full list of restrictions please <a href="https://github.dev/richzhang/colorization/blob/master/LICENSE" target="_blank" style="text-decoration: underline;" target="_blank">read the license</a>
+</p>
+<br>
+<center>
+  <img src='https://visitor-badge.glitch.me/badge?page_id=dj_colorization_eccv16siggraph17' alt='visitor badge'>
+</center>
+"""
+
+#with gr.Interface(css=css) as mainBody:
+gr.HTML("""<style>""" + css+ """</Style>""")
+  
+mainBody = gr.Interface(
+    interface,
+    [
+        gr.components.Image(type="pil", label="image"),
+        gr.components.Radio(
+            ["eccv16", "siggraph17"],
+            type="value",
+            label="model"
+        )
+    ],
+    [
+        gr.components.Image(label="output")
+    ],
+    #inputs="sketchpad",
+    #outputs="label",
+    theme="huggingface",
+    title=title,
+    description=description,
+    article=article,
+    live=True,
+)
+mainBody.launch()

colorizers/__init__.py

@@ -0,0 +1,6 @@
+
+from .base_color import *
+from .eccv16 import *
+from .siggraph17 import *
+from .util import *
+

colorizers/base_color.py

@@ -0,0 +1,24 @@
+
+import torch
+from torch import nn
+
+class BaseColor(nn.Module):
+	def __init__(self):
+		super(BaseColor, self).__init__()
+
+		self.l_cent = 50.
+		self.l_norm = 100.
+		self.ab_norm = 110.
+
+	def normalize_l(self, in_l):
+		return (in_l-self.l_cent)/self.l_norm
+
+	def unnormalize_l(self, in_l):
+		return in_l*self.l_norm + self.l_cent
+
+	def normalize_ab(self, in_ab):
+		return in_ab/self.ab_norm
+
+	def unnormalize_ab(self, in_ab):
+		return in_ab*self.ab_norm
+

colorizers/eccv16.py

@@ -0,0 +1,105 @@
+
+import torch
+import torch.nn as nn
+import numpy as np
+from IPython import embed
+
+from .base_color import *
+
+class ECCVGenerator(BaseColor):
+    def __init__(self, norm_layer=nn.BatchNorm2d):
+        super(ECCVGenerator, self).__init__()
+
+        model1=[nn.Conv2d(1, 64, kernel_size=3, stride=1, padding=1, bias=True),]
+        model1+=[nn.ReLU(True),]
+        model1+=[nn.Conv2d(64, 64, kernel_size=3, stride=2, padding=1, bias=True),]
+        model1+=[nn.ReLU(True),]
+        model1+=[norm_layer(64),]
+
+        model2=[nn.Conv2d(64, 128, kernel_size=3, stride=1, padding=1, bias=True),]
+        model2+=[nn.ReLU(True),]
+        model2+=[nn.Conv2d(128, 128, kernel_size=3, stride=2, padding=1, bias=True),]
+        model2+=[nn.ReLU(True),]
+        model2+=[norm_layer(128),]
+
+        model3=[nn.Conv2d(128, 256, kernel_size=3, stride=1, padding=1, bias=True),]
+        model3+=[nn.ReLU(True),]
+        model3+=[nn.Conv2d(256, 256, kernel_size=3, stride=1, padding=1, bias=True),]
+        model3+=[nn.ReLU(True),]
+        model3+=[nn.Conv2d(256, 256, kernel_size=3, stride=2, padding=1, bias=True),]
+        model3+=[nn.ReLU(True),]
+        model3+=[norm_layer(256),]
+
+        model4=[nn.Conv2d(256, 512, kernel_size=3, stride=1, padding=1, bias=True),]
+        model4+=[nn.ReLU(True),]
+        model4+=[nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1, bias=True),]
+        model4+=[nn.ReLU(True),]
+        model4+=[nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1, bias=True),]
+        model4+=[nn.ReLU(True),]
+        model4+=[norm_layer(512),]
+
+        model5=[nn.Conv2d(512, 512, kernel_size=3, dilation=2, stride=1, padding=2, bias=True),]
+        model5+=[nn.ReLU(True),]
+        model5+=[nn.Conv2d(512, 512, kernel_size=3, dilation=2, stride=1, padding=2, bias=True),]
+        model5+=[nn.ReLU(True),]
+        model5+=[nn.Conv2d(512, 512, kernel_size=3, dilation=2, stride=1, padding=2, bias=True),]
+        model5+=[nn.ReLU(True),]
+        model5+=[norm_layer(512),]
+
+        model6=[nn.Conv2d(512, 512, kernel_size=3, dilation=2, stride=1, padding=2, bias=True),]
+        model6+=[nn.ReLU(True),]
+        model6+=[nn.Conv2d(512, 512, kernel_size=3, dilation=2, stride=1, padding=2, bias=True),]
+        model6+=[nn.ReLU(True),]
+        model6+=[nn.Conv2d(512, 512, kernel_size=3, dilation=2, stride=1, padding=2, bias=True),]
+        model6+=[nn.ReLU(True),]
+        model6+=[norm_layer(512),]
+
+        model7=[nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1, bias=True),]
+        model7+=[nn.ReLU(True),]
+        model7+=[nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1, bias=True),]
+        model7+=[nn.ReLU(True),]
+        model7+=[nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1, bias=True),]
+        model7+=[nn.ReLU(True),]
+        model7+=[norm_layer(512),]
+
+        model8=[nn.ConvTranspose2d(512, 256, kernel_size=4, stride=2, padding=1, bias=True),]
+        model8+=[nn.ReLU(True),]
+        model8+=[nn.Conv2d(256, 256, kernel_size=3, stride=1, padding=1, bias=True),]
+        model8+=[nn.ReLU(True),]
+        model8+=[nn.Conv2d(256, 256, kernel_size=3, stride=1, padding=1, bias=True),]
+        model8+=[nn.ReLU(True),]
+
+        model8+=[nn.Conv2d(256, 313, kernel_size=1, stride=1, padding=0, bias=True),]
+
+        self.model1 = nn.Sequential(*model1)
+        self.model2 = nn.Sequential(*model2)
+        self.model3 = nn.Sequential(*model3)
+        self.model4 = nn.Sequential(*model4)
+        self.model5 = nn.Sequential(*model5)
+        self.model6 = nn.Sequential(*model6)
+        self.model7 = nn.Sequential(*model7)
+        self.model8 = nn.Sequential(*model8)
+
+        self.softmax = nn.Softmax(dim=1)
+        self.model_out = nn.Conv2d(313, 2, kernel_size=1, padding=0, dilation=1, stride=1, bias=False)
+        self.upsample4 = nn.Upsample(scale_factor=4, mode='bilinear')
+
+    def forward(self, input_l):
+        conv1_2 = self.model1(self.normalize_l(input_l))
+        conv2_2 = self.model2(conv1_2)
+        conv3_3 = self.model3(conv2_2)
+        conv4_3 = self.model4(conv3_3)
+        conv5_3 = self.model5(conv4_3)
+        conv6_3 = self.model6(conv5_3)
+        conv7_3 = self.model7(conv6_3)
+        conv8_3 = self.model8(conv7_3)
+        out_reg = self.model_out(self.softmax(conv8_3))
+
+        return self.unnormalize_ab(self.upsample4(out_reg))
+
+def eccv16(pretrained=True):
+	model = ECCVGenerator()
+	if(pretrained):
+		import torch.utils.model_zoo as model_zoo
+		model.load_state_dict(model_zoo.load_url('https://colorizers.s3.us-east-2.amazonaws.com/colorization_release_v2-9b330a0b.pth',map_location='cpu',check_hash=True))
+	return model

colorizers/siggraph17.py

@@ -0,0 +1,168 @@
+import torch
+import torch.nn as nn
+
+from .base_color import *
+
+class SIGGRAPHGenerator(BaseColor):
+    def __init__(self, norm_layer=nn.BatchNorm2d, classes=529):
+        super(SIGGRAPHGenerator, self).__init__()
+
+        # Conv1
+        model1=[nn.Conv2d(4, 64, kernel_size=3, stride=1, padding=1, bias=True),]
+        model1+=[nn.ReLU(True),]
+        model1+=[nn.Conv2d(64, 64, kernel_size=3, stride=1, padding=1, bias=True),]
+        model1+=[nn.ReLU(True),]
+        model1+=[norm_layer(64),]
+        # add a subsampling operation
+
+        # Conv2
+        model2=[nn.Conv2d(64, 128, kernel_size=3, stride=1, padding=1, bias=True),]
+        model2+=[nn.ReLU(True),]
+        model2+=[nn.Conv2d(128, 128, kernel_size=3, stride=1, padding=1, bias=True),]
+        model2+=[nn.ReLU(True),]
+        model2+=[norm_layer(128),]
+        # add a subsampling layer operation
+
+        # Conv3
+        model3=[nn.Conv2d(128, 256, kernel_size=3, stride=1, padding=1, bias=True),]
+        model3+=[nn.ReLU(True),]
+        model3+=[nn.Conv2d(256, 256, kernel_size=3, stride=1, padding=1, bias=True),]
+        model3+=[nn.ReLU(True),]
+        model3+=[nn.Conv2d(256, 256, kernel_size=3, stride=1, padding=1, bias=True),]
+        model3+=[nn.ReLU(True),]
+        model3+=[norm_layer(256),]
+        # add a subsampling layer operation
+
+        # Conv4
+        model4=[nn.Conv2d(256, 512, kernel_size=3, stride=1, padding=1, bias=True),]
+        model4+=[nn.ReLU(True),]
+        model4+=[nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1, bias=True),]
+        model4+=[nn.ReLU(True),]
+        model4+=[nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1, bias=True),]
+        model4+=[nn.ReLU(True),]
+        model4+=[norm_layer(512),]
+
+        # Conv5
+        model5=[nn.Conv2d(512, 512, kernel_size=3, dilation=2, stride=1, padding=2, bias=True),]
+        model5+=[nn.ReLU(True),]
+        model5+=[nn.Conv2d(512, 512, kernel_size=3, dilation=2, stride=1, padding=2, bias=True),]
+        model5+=[nn.ReLU(True),]
+        model5+=[nn.Conv2d(512, 512, kernel_size=3, dilation=2, stride=1, padding=2, bias=True),]
+        model5+=[nn.ReLU(True),]
+        model5+=[norm_layer(512),]
+
+        # Conv6
+        model6=[nn.Conv2d(512, 512, kernel_size=3, dilation=2, stride=1, padding=2, bias=True),]
+        model6+=[nn.ReLU(True),]
+        model6+=[nn.Conv2d(512, 512, kernel_size=3, dilation=2, stride=1, padding=2, bias=True),]
+        model6+=[nn.ReLU(True),]
+        model6+=[nn.Conv2d(512, 512, kernel_size=3, dilation=2, stride=1, padding=2, bias=True),]
+        model6+=[nn.ReLU(True),]
+        model6+=[norm_layer(512),]
+
+        # Conv7
+        model7=[nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1, bias=True),]
+        model7+=[nn.ReLU(True),]
+        model7+=[nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1, bias=True),]
+        model7+=[nn.ReLU(True),]
+        model7+=[nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1, bias=True),]
+        model7+=[nn.ReLU(True),]
+        model7+=[norm_layer(512),]
+
+        # Conv7
+        model8up=[nn.ConvTranspose2d(512, 256, kernel_size=4, stride=2, padding=1, bias=True)]
+        model3short8=[nn.Conv2d(256, 256, kernel_size=3, stride=1, padding=1, bias=True),]
+
+        model8=[nn.ReLU(True),]
+        model8+=[nn.Conv2d(256, 256, kernel_size=3, stride=1, padding=1, bias=True),]
+        model8+=[nn.ReLU(True),]
+        model8+=[nn.Conv2d(256, 256, kernel_size=3, stride=1, padding=1, bias=True),]
+        model8+=[nn.ReLU(True),]
+        model8+=[norm_layer(256),]
+
+        # Conv9
+        model9up=[nn.ConvTranspose2d(256, 128, kernel_size=4, stride=2, padding=1, bias=True),]
+        model2short9=[nn.Conv2d(128, 128, kernel_size=3, stride=1, padding=1, bias=True),]
+        # add the two feature maps above        
+
+        model9=[nn.ReLU(True),]
+        model9+=[nn.Conv2d(128, 128, kernel_size=3, stride=1, padding=1, bias=True),]
+        model9+=[nn.ReLU(True),]
+        model9+=[norm_layer(128),]
+
+        # Conv10
+        model10up=[nn.ConvTranspose2d(128, 128, kernel_size=4, stride=2, padding=1, bias=True),]
+        model1short10=[nn.Conv2d(64, 128, kernel_size=3, stride=1, padding=1, bias=True),]
+        # add the two feature maps above
+
+        model10=[nn.ReLU(True),]
+        model10+=[nn.Conv2d(128, 128, kernel_size=3, dilation=1, stride=1, padding=1, bias=True),]
+        model10+=[nn.LeakyReLU(negative_slope=.2),]
+
+        # classification output
+        model_class=[nn.Conv2d(256, classes, kernel_size=1, padding=0, dilation=1, stride=1, bias=True),]
+
+        # regression output
+        model_out=[nn.Conv2d(128, 2, kernel_size=1, padding=0, dilation=1, stride=1, bias=True),]
+        model_out+=[nn.Tanh()]
+
+        self.model1 = nn.Sequential(*model1)
+        self.model2 = nn.Sequential(*model2)
+        self.model3 = nn.Sequential(*model3)
+        self.model4 = nn.Sequential(*model4)
+        self.model5 = nn.Sequential(*model5)
+        self.model6 = nn.Sequential(*model6)
+        self.model7 = nn.Sequential(*model7)
+        self.model8up = nn.Sequential(*model8up)
+        self.model8 = nn.Sequential(*model8)
+        self.model9up = nn.Sequential(*model9up)
+        self.model9 = nn.Sequential(*model9)
+        self.model10up = nn.Sequential(*model10up)
+        self.model10 = nn.Sequential(*model10)
+        self.model3short8 = nn.Sequential(*model3short8)
+        self.model2short9 = nn.Sequential(*model2short9)
+        self.model1short10 = nn.Sequential(*model1short10)
+
+        self.model_class = nn.Sequential(*model_class)
+        self.model_out = nn.Sequential(*model_out)
+
+        self.upsample4 = nn.Sequential(*[nn.Upsample(scale_factor=4, mode='bilinear'),])
+        self.softmax = nn.Sequential(*[nn.Softmax(dim=1),])
+
+    def forward(self, input_A, input_B=None, mask_B=None):
+        if(input_B is None):
+            input_B = torch.cat((input_A*0, input_A*0), dim=1)
+        if(mask_B is None):
+            mask_B = input_A*0
+
+        conv1_2 = self.model1(torch.cat((self.normalize_l(input_A),self.normalize_ab(input_B),mask_B),dim=1))
+        conv2_2 = self.model2(conv1_2[:,:,::2,::2])
+        conv3_3 = self.model3(conv2_2[:,:,::2,::2])
+        conv4_3 = self.model4(conv3_3[:,:,::2,::2])
+        conv5_3 = self.model5(conv4_3)
+        conv6_3 = self.model6(conv5_3)
+        conv7_3 = self.model7(conv6_3)
+
+        conv8_up = self.model8up(conv7_3) + self.model3short8(conv3_3)
+        conv8_3 = self.model8(conv8_up)
+        conv9_up = self.model9up(conv8_3) + self.model2short9(conv2_2)
+        conv9_3 = self.model9(conv9_up)
+        conv10_up = self.model10up(conv9_3) + self.model1short10(conv1_2)
+        conv10_2 = self.model10(conv10_up)
+        out_reg = self.model_out(conv10_2)
+
+        conv9_up = self.model9up(conv8_3) + self.model2short9(conv2_2)
+        conv9_3 = self.model9(conv9_up)
+        conv10_up = self.model10up(conv9_3) + self.model1short10(conv1_2)
+        conv10_2 = self.model10(conv10_up)
+        out_reg = self.model_out(conv10_2)
+
+        return self.unnormalize_ab(out_reg)
+
+def siggraph17(pretrained=True):
+    model = SIGGRAPHGenerator()
+    if(pretrained):
+        import torch.utils.model_zoo as model_zoo
+        model.load_state_dict(model_zoo.load_url('https://colorizers.s3.us-east-2.amazonaws.com/siggraph17-df00044c.pth',map_location='cpu',check_hash=True))
+    return model
+

colorizers/util.py

@@ -0,0 +1,47 @@
+
+from PIL import Image
+import numpy as np
+from skimage import color
+import torch
+import torch.nn.functional as F
+from IPython import embed
+
+def load_img(img_path):
+	out_np = np.asarray(Image.open(img_path))
+	if(out_np.ndim==2):
+		out_np = np.tile(out_np[:,:,None],3)
+	return out_np
+
+def resize_img(img, HW=(256,256), resample=3):
+	return np.asarray(Image.fromarray(img).resize((HW[1],HW[0]), resample=resample))
+
+def preprocess_img(img_rgb_orig, HW=(256,256), resample=3):
+	# return original size L and resized L as torch Tensors
+	img_rgb_rs = resize_img(img_rgb_orig, HW=HW, resample=resample)
+	
+	img_lab_orig = color.rgb2lab(img_rgb_orig)
+	img_lab_rs = color.rgb2lab(img_rgb_rs)
+
+	img_l_orig = img_lab_orig[:,:,0]
+	img_l_rs = img_lab_rs[:,:,0]
+
+	tens_orig_l = torch.Tensor(img_l_orig)[None,None,:,:]
+	tens_rs_l = torch.Tensor(img_l_rs)[None,None,:,:]
+
+	return (tens_orig_l, tens_rs_l)
+
+def postprocess_tens(tens_orig_l, out_ab, mode='bilinear'):
+	# tens_orig_l 	1 x 1 x H_orig x W_orig
+	# out_ab 		1 x 2 x H x W
+
+	HW_orig = tens_orig_l.shape[2:]
+	HW = out_ab.shape[2:]
+
+	# call resize function if needed
+	if(HW_orig[0]!=HW[0] or HW_orig[1]!=HW[1]):
+		out_ab_orig = F.interpolate(out_ab, size=HW_orig, mode='bilinear')
+	else:
+		out_ab_orig = out_ab
+
+	out_lab_orig = torch.cat((tens_orig_l, out_ab_orig), dim=1)
+	return color.lab2rgb(out_lab_orig.data.cpu().numpy()[0,...].transpose((1,2,0)))

requirements.txt

@@ -0,0 +1,5 @@
+torch
+scikit-image
+numpy
+matplotlib
+IPython