An unofficial implementation of View Independent Generative Adversarial Network for Novel View Synthesis.
Target | ![]() |
![]() |
![]() |
---|---|---|---|
Predict | ![]() |
![]() |
![]() |
- pytorch 1.5.1
- torchvision 0.6.1
python train.py --train_dir [/folder/to/training/dataset] --batch_size [] --checkpoint [./checkpoint/0.pth]
python test.py --test_dir [/folder/to/testing/dataset] --checkpoint [./checkpoint/x.pth]
The data is based on the Shapenet dataset naming convention: {model name}_{azimuth/10}_{elevation}.png
[1] Xu, Xiaogang, Ying-Cong Chen, and Jiaya Jia. "View Independent Generative Adversarial Network for Novel View Synthesis." Proceedings of the IEEE International Conference on Computer Vision. 2019.