Apply the model on a interactive art demo

[HarryGuan22]

Thank you so much for sharing your artistic model to the community. I'm a media art student new to machine learning. I'm currently apply your 'yellow comic alien 004 kimg' model on a interactive art demo. It looks pretty cool and I'm really appriciate your time and effort!
Here is a video preview: https://youtu.be/lVJnj5PLghs

[edstoica]

hey harry,
thanks for the message.
so cool to see other people using the datasets and models.
thanks also to include the link in you video description 👍

your idea is really funny. may you explain to me: how does posenet interact with the model?

[HarryGuan22]

It's very nice to hear from you. I just set up OSC communication between visualizer.py and the ml5.js PoseNet. In visualizer.py, I integrate python.osc to make it available to receive information via OSC(open sound control) in real-time. And then I apply osc.js to send ml5.js PoseNet captured data to the specific OSC port. On the visualizer.py side, there is a dragging behavior to walk through the latent space. I just replaced the changing mouse position (dx,dy) with the OSC data from ml5.js PoseNet. Also, you can do something to other parameters like Truncate. Hope this helps! lucid layers ***@***.***> 于2022年2月21日周一 10:12写道：

…

Closed #1 <#1>. — Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ARSR6KK7CH26NDKMQ6E5GUDU4GNSRANCNFSM5OTWHSUA> . You are receiving this because you authored the thread.Message ID: <edstoica/lucid_stylegan3_datasets_models/issue/1/issue_event/6108308000@ github.com>