

GET3D (Nvidia)
3D

Some of the recent works of NV-TLabs include:
- Neural Radiance Fields (NeRF), a method that uses deep neural networks to represent 3D scenes as continuous functions of 3D coordinates and viewing directions, and enables high-quality novel view synthesis from sparse and unstructured images.
- StyleGAN2-ADA, an improved version of StyleGAN2 that adapts to different data domains using adaptive discriminator augmentation, and achieves state-of-the-art results in generating realistic and diverse images of faces, cars, cats, and more.
- GANcraft, a system that converts block-based 3D models (such as those created in Minecraft) into realistic scenes using generative adversarial networks (GANs), and allows users to interactively edit and explore the generated worlds.
NV-TLabs also maintains a website where they showcase their research outputs, publications, code, and datasets. The website is https://nv-tlabs.github.io/, and it is updated regularly with new content and features. The website also provides links to the social media accounts of NV-TLabs, where they share news and updates about their work.
If you are interested in learning more about NV-TLabs and their research, you can visit their website or follow them on Twitter (@NVTlabs) or YouTube (NVIDIA Research). You can also contact them via email at nv-tlabs@nvidia.com.
Pros
- It showcases the research projects and publications of NVIDIA Toronto AI Lab, which is a leading group in computer vision and machine learning .
- It provides code and data for some of the projects, such as FEGR, GENIE, LION, and GET3D , which can be useful for researchers and practitioners who want to reproduce or build upon the work.
- It has a simple and elegant design that is easy to navigate and browse .
Cons
- It does not have a blog or news section that updates the latest developments and events of the lab .
- It does not have a contact or feedback form that allows visitors to communicate with the lab members or ask questions .
- It does not have a clear license or terms of use for the code and data that are provided, which may cause confusion or legal issues for the users .