Spaces:
Build error
Build error
updated app
Browse files
app.py
CHANGED
@@ -1,29 +1,25 @@
|
|
1 |
-
import
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
import torch
|
3 |
-
import argparse
|
4 |
-
|
5 |
-
import numpy as np
|
6 |
import torch
|
7 |
-
import matplotlib.pyplot as plt
|
8 |
-
from PIL import Image
|
9 |
-
|
10 |
-
print(torch.__version__)
|
11 |
-
print(torch.version.cuda)
|
12 |
|
13 |
sys.path.append("./rome/")
|
14 |
sys.path.append('./DECA')
|
15 |
|
16 |
-
from rome.src.utils import args as args_utils
|
17 |
-
from rome.src.utils.processing import process_black_shape, tensor2image
|
18 |
|
19 |
# loading models ---- create model repo
|
20 |
-
from huggingface_hub import hf_hub_download
|
21 |
|
22 |
-
default_modnet_path = hf_hub_download(
|
23 |
-
|
|
|
24 |
|
25 |
# parser configurations
|
26 |
-
from easydict import EasyDict as edict
|
27 |
|
28 |
args = edict({
|
29 |
"save_dir": ".",
|
@@ -93,8 +89,8 @@ args = edict({
|
|
93 |
"num_vertex": 5023,
|
94 |
"train_basis": True,
|
95 |
"path_to_deca": "DECA",
|
96 |
-
"path_to_linear_hair_model": "data/linear_hair.pth",
|
97 |
-
"path_to_mobile_model": "data/disp_model.pth",
|
98 |
"n_scalp": 60,
|
99 |
"use_distill": False,
|
100 |
"use_mobile_version": False,
|
@@ -109,40 +105,49 @@ args = edict({
|
|
109 |
})
|
110 |
|
111 |
# download FLAME and DECA pretrained
|
112 |
-
generic_model_path = hf_hub_download('Pie31415/rome','generic_model.pkl')
|
113 |
-
deca_model_path = hf_hub_download('Pie31415/rome','deca_model.tar')
|
114 |
-
|
115 |
-
import pickle
|
116 |
|
117 |
with open(generic_model_path, 'rb') as f:
|
118 |
-
|
119 |
|
120 |
-
|
121 |
-
|
122 |
|
123 |
with open(deca_model_path, "rb") as input:
|
124 |
-
|
125 |
-
|
126 |
-
|
127 |
|
128 |
# load ROME inference model
|
129 |
-
from rome.infer import Infer
|
130 |
infer = Infer(args)
|
131 |
|
132 |
-
def
|
|
|
|
|
|
|
133 |
out = infer.evaluate(source_img, driver_img, crop_center=False)
|
134 |
res = tensor2image(torch.cat([out['source_information']['data_dict']['source_img'][0].cpu(),
|
135 |
-
|
136 |
-
|
|
|
137 |
return res[..., ::-1]
|
138 |
|
139 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
140 |
|
141 |
-
|
142 |
-
fn=predict,
|
143 |
-
inputs=[
|
144 |
-
gr.Image(type="pil"),
|
145 |
-
gr.Image(type="pil")
|
146 |
-
],
|
147 |
-
outputs=gr.Image(),
|
148 |
-
examples=[]).launch()
|
|
|
1 |
+
import gradio as gr
|
2 |
+
from rome.infer import Infer
|
3 |
+
import pickle
|
4 |
+
from easydict import EasyDict as edict
|
5 |
+
from huggingface_hub import hf_hub_download
|
6 |
+
from rome.src.utils.processing import process_black_shape, tensor2image
|
7 |
+
from rome.src.utils import args as args_utils
|
8 |
+
import sys
|
9 |
import torch
|
|
|
|
|
|
|
10 |
import torch
|
|
|
|
|
|
|
|
|
|
|
11 |
|
12 |
sys.path.append("./rome/")
|
13 |
sys.path.append('./DECA')
|
14 |
|
|
|
|
|
15 |
|
16 |
# loading models ---- create model repo
|
|
|
17 |
|
18 |
+
default_modnet_path = hf_hub_download(
|
19 |
+
'Pie31415/rome', 'modnet_photographic_portrait_matting.ckpt')
|
20 |
+
default_model_path = hf_hub_download('Pie31415/rome', 'rome.pth')
|
21 |
|
22 |
# parser configurations
|
|
|
23 |
|
24 |
args = edict({
|
25 |
"save_dir": ".",
|
|
|
89 |
"num_vertex": 5023,
|
90 |
"train_basis": True,
|
91 |
"path_to_deca": "DECA",
|
92 |
+
"path_to_linear_hair_model": "data/linear_hair.pth", # N/A
|
93 |
+
"path_to_mobile_model": "data/disp_model.pth", # N/A
|
94 |
"n_scalp": 60,
|
95 |
"use_distill": False,
|
96 |
"use_mobile_version": False,
|
|
|
105 |
})
|
106 |
|
107 |
# download FLAME and DECA pretrained
|
108 |
+
generic_model_path = hf_hub_download('Pie31415/rome', 'generic_model.pkl')
|
109 |
+
deca_model_path = hf_hub_download('Pie31415/rome', 'deca_model.tar')
|
|
|
|
|
110 |
|
111 |
with open(generic_model_path, 'rb') as f:
|
112 |
+
ss = pickle.load(f, encoding='latin1')
|
113 |
|
114 |
+
with open('./DECA/data/generic_model.pkl', 'wb') as out:
|
115 |
+
pickle.dump(ss, out)
|
116 |
|
117 |
with open(deca_model_path, "rb") as input:
|
118 |
+
with open('./DECA/data/deca_model.tar', "wb") as out:
|
119 |
+
for line in input:
|
120 |
+
out.write(line)
|
121 |
|
122 |
# load ROME inference model
|
|
|
123 |
infer = Infer(args)
|
124 |
|
125 |
+
def image_inference(
|
126 |
+
source_img: gr.inputs.Image = None,
|
127 |
+
driver_img: gr.inputs.Image = None
|
128 |
+
):
|
129 |
out = infer.evaluate(source_img, driver_img, crop_center=False)
|
130 |
res = tensor2image(torch.cat([out['source_information']['data_dict']['source_img'][0].cpu(),
|
131 |
+
out['source_information']['data_dict']['target_img'][0].cpu(
|
132 |
+
),
|
133 |
+
out['render_masked'].cpu(), out['pred_target_shape_img'][0].cpu()], dim=2))
|
134 |
return res[..., ::-1]
|
135 |
|
136 |
+
def folder_inference():
|
137 |
+
pass
|
138 |
+
|
139 |
+
with gr.Blocks() as demo:
|
140 |
+
with gr.Tab("Image Inference"):
|
141 |
+
image_input = [gr.Image(type="pil"), gr.Image(type="pil")]
|
142 |
+
image_output = gr.Image()
|
143 |
+
image_button = gr.Button("Predict")
|
144 |
+
with gr.Tab("Inference Over Folder"):
|
145 |
+
pass
|
146 |
+
with gr.Tab("Video Inference"):
|
147 |
+
pass
|
148 |
+
|
149 |
+
image_button.click(image_inference, inputs=image_input, outputs=image_output)
|
150 |
+
title = "ROME: Realistic one-shot mesh-based head avatars"
|
151 |
+
examples = gr.Examples(["examples/lincoln.jpg", "examples/tars2.jpg"])
|
152 |
|
153 |
+
demo.launch()
|
|
|
|
|
|
|
|
|
|
|
|
|
|