Spaces:
Build error
Build error
updated app
Browse files
app.py
CHANGED
@@ -13,9 +13,7 @@ sys.path.append("./rome/")
|
|
13 |
sys.path.append('./DECA')
|
14 |
|
15 |
# loading models ---- create model repo
|
16 |
-
|
17 |
-
default_modnet_path = hf_hub_download(
|
18 |
-
'Pie31415/rome', 'modnet_photographic_portrait_matting.ckpt')
|
19 |
default_model_path = hf_hub_download('Pie31415/rome', 'rome.pth')
|
20 |
|
21 |
# parser configurations
|
@@ -126,25 +124,23 @@ def image_inference(
|
|
126 |
):
|
127 |
out = infer.evaluate(source_img, driver_img, crop_center=False)
|
128 |
res = tensor2image(torch.cat([out['source_information']['data_dict']['source_img'][0].cpu(),
|
129 |
-
out['source_information']['data_dict']['target_img'][0].cpu(
|
130 |
-
|
131 |
-
out['render_masked'].cpu(), out['pred_target_shape_img'][0].cpu()], dim=2))
|
132 |
return res[..., ::-1]
|
133 |
|
134 |
-
def
|
135 |
pass
|
136 |
|
137 |
with gr.Blocks() as demo:
|
138 |
with gr.Tab("Image Inference"):
|
139 |
-
|
140 |
image_output = gr.Image()
|
141 |
image_button = gr.Button("Predict")
|
142 |
-
with gr.Tab("Inference Over Folder"):
|
143 |
-
pass
|
144 |
with gr.Tab("Video Inference"):
|
|
|
145 |
pass
|
146 |
|
147 |
-
image_button.click(image_inference, inputs=
|
148 |
title = "ROME: Realistic one-shot mesh-based head avatars"
|
149 |
examples = gr.Examples(["examples/lincoln.jpg", "examples/tars2.jpg"])
|
150 |
|
|
|
13 |
sys.path.append('./DECA')
|
14 |
|
15 |
# loading models ---- create model repo
|
16 |
+
default_modnet_path = hf_hub_download('Pie31415/rome', 'modnet_photographic_portrait_matting.ckpt')
|
|
|
|
|
17 |
default_model_path = hf_hub_download('Pie31415/rome', 'rome.pth')
|
18 |
|
19 |
# parser configurations
|
|
|
124 |
):
|
125 |
out = infer.evaluate(source_img, driver_img, crop_center=False)
|
126 |
res = tensor2image(torch.cat([out['source_information']['data_dict']['source_img'][0].cpu(),
|
127 |
+
out['source_information']['data_dict']['target_img'][0].cpu(),
|
128 |
+
out['render_masked'].cpu(), out['pred_target_shape_img'][0].cpu()], dim=2))
|
|
|
129 |
return res[..., ::-1]
|
130 |
|
131 |
+
def video_inference():
|
132 |
pass
|
133 |
|
134 |
with gr.Blocks() as demo:
|
135 |
with gr.Tab("Image Inference"):
|
136 |
+
image_inputs = [gr.Image(type="pil"), gr.Image(type="pil")]
|
137 |
image_output = gr.Image()
|
138 |
image_button = gr.Button("Predict")
|
|
|
|
|
139 |
with gr.Tab("Video Inference"):
|
140 |
+
video_inputs = [gr.Video(), gr.Image()]
|
141 |
pass
|
142 |
|
143 |
+
image_button.click(image_inference, inputs=image_inputs, outputs=image_output)
|
144 |
title = "ROME: Realistic one-shot mesh-based head avatars"
|
145 |
examples = gr.Examples(["examples/lincoln.jpg", "examples/tars2.jpg"])
|
146 |
|