Spaces:
Runtime error
Runtime error
RamAnanth1
commited on
Commit
•
ed36721
1
Parent(s):
ee7d022
Include pointcloud2mesh components to additionally view 3d model
Browse filesAdd pointcloud2mesh components from the notebook (https://github.com/openai/point-e/blob/main/point_e/examples/pointcloud2mesh.ipynb)
app.py
CHANGED
@@ -9,6 +9,9 @@ from point_e.diffusion.sampler import PointCloudSampler
|
|
9 |
from point_e.models.download import load_checkpoint
|
10 |
from point_e.models.configs import MODEL_CONFIGS, model_from_config
|
11 |
from point_e.util.plotting import plot_point_cloud
|
|
|
|
|
|
|
12 |
|
13 |
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
|
14 |
|
@@ -29,6 +32,14 @@ base_model.load_state_dict(load_checkpoint(base_name, device))
|
|
29 |
print('downloading upsampler checkpoint...')
|
30 |
upsampler_model.load_state_dict(load_checkpoint('upsample', device))
|
31 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
32 |
sampler = PointCloudSampler(
|
33 |
device=device,
|
34 |
models=[base_model, upsampler_model],
|
@@ -65,12 +76,30 @@ def inference(prompt):
|
|
65 |
)
|
66 |
),
|
67 |
)
|
68 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
69 |
|
70 |
demo = gr.Interface(
|
71 |
fn=inference,
|
72 |
inputs="text",
|
73 |
-
outputs=gr.Plot(),
|
74 |
examples=[
|
75 |
["a red motorcycle"],
|
76 |
["a RED pumpkin"],
|
|
|
9 |
from point_e.models.download import load_checkpoint
|
10 |
from point_e.models.configs import MODEL_CONFIGS, model_from_config
|
11 |
from point_e.util.plotting import plot_point_cloud
|
12 |
+
from point_e.util.pc_to_mesh import marching_cubes_mesh
|
13 |
+
|
14 |
+
import trimesh
|
15 |
|
16 |
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
|
17 |
|
|
|
32 |
print('downloading upsampler checkpoint...')
|
33 |
upsampler_model.load_state_dict(load_checkpoint('upsample', device))
|
34 |
|
35 |
+
print('creating SDF model...')
|
36 |
+
name = 'sdf'
|
37 |
+
sdf_model = model_from_config(MODEL_CONFIGS[name], device)
|
38 |
+
sdf_model.eval()
|
39 |
+
|
40 |
+
print('loading SDF model...')
|
41 |
+
sdf_model.load_state_dict(load_checkpoint(name, device))
|
42 |
+
|
43 |
sampler = PointCloudSampler(
|
44 |
device=device,
|
45 |
models=[base_model, upsampler_model],
|
|
|
76 |
)
|
77 |
),
|
78 |
)
|
79 |
+
|
80 |
+
# Produce a mesh (with vertex colors)
|
81 |
+
mesh = marching_cubes_mesh(
|
82 |
+
pc=pc,
|
83 |
+
model=sdf_model,
|
84 |
+
batch_size=4096,
|
85 |
+
grid_size=32, # increase to 128 for resolution used in evals
|
86 |
+
progress=True,
|
87 |
+
)
|
88 |
+
|
89 |
+
# Write the mesh to a PLY file to import into some other program.
|
90 |
+
with open("mesh.ply", 'wb') as f:
|
91 |
+
mesh.write_ply(f)
|
92 |
+
|
93 |
+
obj_file = '3d_model.obj'
|
94 |
+
mesh = trimesh.load('mesh.ply')
|
95 |
+
mesh.export(obj_file)
|
96 |
+
|
97 |
+
return fig, obj_file
|
98 |
|
99 |
demo = gr.Interface(
|
100 |
fn=inference,
|
101 |
inputs="text",
|
102 |
+
outputs=[gr.Plot(),gr.Model3D(value=None)],
|
103 |
examples=[
|
104 |
["a red motorcycle"],
|
105 |
["a RED pumpkin"],
|