When using imto3dv2.py speed is slow when calling api maybe due to internet speed of the server or the limit of free sup get huggingface pro may reduce the generate time.
for greatest performance please use imto3dv2.ipynb
or https://www.kaggle.com/code/minhtranv/image-to-3d-web-app-with-gradio-and-ngrok
requires a token generated from, remember to gain READ in token setting: https://huggingface.co/settings/tokens also u need to fill in the access form in this link : https://huggingface.co/stabilityai/stable-fast-3d
Following step in
if colab runout of compute unit then use kaggle : https://www.kaggle.com/code/minhtranv/image-to-3d-web-app-with-gradio-and-ngrok
While runing on kaggle. Kaggle block direct web deploy on thier server so we need to use ngrok to create a tunnel link to our local host on kaggle server You will access app via tunel link. Remember to read instruction in kaggle
On kaggle app will be access via the public link
app should look like this
create a folder named: Github , then u need to set the clone directory in to that folder.
which should look like that : ./Github
then clone the repo to github folder u created before
final link should look like : ./Github/9.5AI_techday
library require 1 : gradio 2 : Opencv
Finally run imto3dv2.py and the app will deploy in the local link which will show in your terminal.