mingwuzheng / imface Goto Github PK

Really great work! Was just wondering how to replicate this for custom data? (our own monocular videos)

:)

In youtube video, you show that ImFace can control expressions and identities independently. But I never find the code about it. I guess we can change expression/identity embedding to edit attributes. Is that right?

Dear Mingwu,

Congratulation for your great work.
The Fig.1 in your paper is really fancy, could you please teach me how to plot SDF(different level sets) as shown in Fig.3.

Best,
Runsong

Hi, I would like to ask when will the code be open sourced?

请问选择哪个facescape数据集

Hi,

Thanks for your great work. It's really cool. There is one experimental detail that I felt a little confused about. Regarding the Positional Encoding, I noticed in the paper and supplementary materials that it was mentioned as being used, but when I checked the codes, it did not seem to be used during training. I'm not sure if I misunderstood something, so I really look forward to hearing from you and would appreciate it.

Thank you. Have a nice day!

I found that the readme was not cover the usage of the model. Could you explain with more details? Thanks very much.

Hi,

Thanks for your great work!

When I reloaded a model and continued to train, I got some errors:

assert len(optimizer_state["found_inf_per_device"]) > 0, "No inf checks were recorded for this optimizer."
AssertionError: No inf checks were recorded for this optimizer.

Did anyone face the same problem and solve it? Thanks!

Seems like most requirements (torch, torchvision ..etc.) are outdated.
Throws error message when pip install -r "requirement.txt"

Could those be updated, or does someone has a updated list of working requirements?

In your paper, you have specified the 5 landmarks located at the outer eye corners, mouth corners, and nose tip are utilized to describe the local regions. I wonder how to guide the landmark network to focus on these 5 local regions.

Hello, I appreciate your outstanding efforts and the excellent work you've done.
Thank you.
When running the command:
python demo/process.py
as described in README.md, I encounter the following error:
`
/usr/local/lib/python3.8/site-packages/pymeshlab/init.py:41: UserWarning: Deprecated Filter Warning: You should use 'meshing_remove_unreferenced_vertices' instead of 'remove_unreferenced_vertices'. See https://pymeshlab.readthedocs.io/en/latest/index.html#filters-renaming
res_dict = self.apply_filter(name, **kwargs)
libEGL warning: MESA-LOADER: failed to open swrast: /usr/lib/dri/swrast_dri.so: cannot open shared object file: No such file or directory (search paths /usr/lib/x86_64-linux-gnu/dri:$${ORIGIN}/dri:/usr/lib/dri, suffix _dri)

libEGL warning: MESA-LOADER: failed to open swrast: /usr/lib/dri/swrast_dri.so: cannot open shared object file: No such file or directory (search paths /usr/lib/x86_64-linux-gnu/dri:$${ORIGIN}/dri:/usr/lib/dri, suffix _dri)

Traceback (most recent call last):
File "/content/drive/MyDrive/ImFace-master/ImFace-master/data_preprocess/preprocess.py", line 111, in process_normal
surface_points, freespace_points = surface_point_cloud.sample_on_surface(
File "/content/drive/MyDrive/ImFace-master/ImFace-master/data_preprocess/surface_point_cloud.py", line 298, in sample_on_surface
surface_pcl = get_surface_point_cloud(mesh, surface_point_method, 1, scan_count,
File "/content/drive/MyDrive/ImFace-master/ImFace-master/data_preprocess/surface_point_cloud.py", line 210, in get_surface_point_cloud
return create_from_scans(mesh, bounding_radius=bounding_radius, scan_count=scan_count,
File "/content/drive/MyDrive/ImFace-master/ImFace-master/data_preprocess/surface_point_cloud.py", line 233, in create_from_scans
scans.append(Scan(mesh,
File "/usr/local/lib/python3.8/site-packages/mesh_to_sdf/scan.py", line 61, in init
color, depth = render_normal_and_depth_buffers(mesh, camera, self.camera_transform, resolution)
File "/usr/local/lib/python3.8/site-packages/mesh_to_sdf/pyrender_wrapper.py", line 58, in render_normal_and_depth_buffers
renderer = pyrender.OffscreenRenderer(resolution, resolution)
File "/usr/local/lib/python3.8/site-packages/pyrender/offscreen.py", line 31, in init
self._create()
File "/usr/local/lib/python3.8/site-packages/pyrender/offscreen.py", line 149, in _create
self._platform.init_context()
File "/usr/local/lib/python3.8/site-packages/pyrender/platforms/egl.py", line 177, in init_context
assert eglInitialize(self._egl_display, major, minor)
File "/usr/local/lib/python3.8/site-packages/OpenGL/platform/baseplatform.py", line 402, in call
return self( *args, **named )
File "/usr/local/lib/python3.8/site-packages/OpenGL/error.py", line 228, in glCheckError
raise GLError(
OpenGL.error.GLError: GLError(
err = 12289,
baseOperation = eglInitialize,
cArguments = (
<OpenGL._opaque.EGLDisplay_pointer object at 0x7e906074c7c0>,
c_long(0),
c_long(0),
),
result = 0
`
I would be very grateful if you could assist me with this.

Hi, I'm stumbling on this error :

ImportError: cannot import name 'fileio' from 'utils'

Could you tell me where is utils supposed to be ?