no code implementations • 13 Dec 2023 • Wonbong Jang, Lourdes Agapito
We propose NViST, a transformer-based model for efficient and generalizable novel-view synthesis from a single image for real-world scenes.
1 code implementation • ICCV 2021 • Wonbong Jang, Lourdes Agapito
At test time, given a single unposed image of an unseen object, CodeNeRF jointly estimates camera viewpoint, and shape and appearance codes via optimization.
1 code implementation • 27 Dec 2018 • Wonbong Jang
CWGAN-GP tends to generate more diverse images than other models using reconstruction loss.