-
### Description
Hi I have used the facelandmark with blendshapes v2 model in my Vtube app.
The issue is that it runs really slowly.
I have tested everything to see what is causing the slow processi…
-
### Have I written custom code (as opposed to using a stock example script provided in MediaPipe)
Yes
### OS Platform and Distribution
Android Webview Chrome 92
### MediaPipe Tasks SDK version
0.…
-
### Have I written custom code (as opposed to using a stock example script provided in MediaPipe)
None
### OS Platform and Distribution
macOS Sonoma 14.0
### Browser and version if the iss…
-
**Is your feature request related to a problem you are facing?**
I would like to be able to resume speakerstats counter for participants with same id provided via token after reload instead of creati…
-
### Search before asking
- [X] I have searched the YOLOv5 [issues](https://github.com/ultralytics/yolov5/issues) and [discussions](https://github.com/ultralytics/yolov5/discussions) and found no simi…
-
Hello,
I have a complex scikit-learn pipeline which I have been trying to convert to ONNX.
The complexity comes from me trying to "mask out" parts of the input for different components of the pi…
-
Currently I'm making a simple Live2D Mocap tool that runs only on web.
I was able to make the Mocap feature works just fine, but it is blocking main thread, so whenever Motion tracking feature is e…
-
Great work! Thanks for sharing the code.
I have noticed that during the texture fitting process in the final visualization stage (at [https://github.com/youngLBW/HRN/blob/main/models/facerecon_model.…
kizum updated
10 months ago
-
Thanks very much for sharing your work! But when I'm training as the readme introduces, I meet problems below, hope you can help, thank you!(python3.6.1, torch1.2.0)
do:
python train.py
problem:
…
-
### Plugin Version or Commit ID
v0.14.3
### Unity Version
2022.3.27f1c1
### Your Host OS
Windows 10
### Target Platform
UnityEditor
### Description
I would like to express my sincere gratitud…