Skip to content

Latest commit

 

History

History
31 lines (21 loc) · 1.15 KB

README.md

File metadata and controls

31 lines (21 loc) · 1.15 KB
Raw

SGSG: Semantics-Guided Shape Generation

This repository contains the implementation of the label regression network (LRN) and the web application of the semantics-guided shape generator described in the paper "Semantics-Guided Latent Space Exploration for Shape Generation". For the implementations of the shape encoder network (SEN) and the shape decoder network (SDN) also described in the paper, please refer to IsaacGuan/3D-GAE and IsaacGuan/implicit-decoder/IMGAN, respectively.

Training LRN

Run the following commands.

python train.py --dataset_name chairs
python train.py --dataset_name lamps
python train.py --dataset_name tables

Testing LRN and Generating the Gaussians

Run the following commands.

python test.py --dataset_name chairs
python test.py --dataset_name lamps
python test.py --dataset_name tables

Running the Web Application

In the web-app folder, run the following command and then visit http://localhost:5000/ from your browser.

python app.py