Average Ratings 0 Ratings
Average Ratings 0 Ratings
Description
Creating visual content that aligns with user requirements often necessitates a high degree of flexibility and precision in managing the pose, shape, expression, and arrangement of the generated elements. Traditional methods enhance the controllability of generative adversarial networks (GANs) by relying on manually labeled training datasets or pre-existing 3D models, which frequently fall short in terms of flexibility, accuracy, and adaptability. In this research, we explore a powerful yet relatively underutilized technique for controlling GANs, which allows users to "drag" specific points in an image to accurately reach designated target locations through interactive engagement, as illustrated in Fig.1. Our proposed solution, DragGAN, comprises two primary components: first, a feature-based motion supervision system that guides the handle point toward the intended position; and second, an innovative point tracking method that utilizes the discriminative features of GANs to continuously identify the handle points' locations. With DragGAN, users gain the capability to manipulate images with exceptional precision in directing pixel movements, thereby facilitating a more intuitive and user-centered design process. This approach not only enhances creative possibilities but also empowers users to achieve their desired visual outcomes more effectively.
Description
We create a three-dimensional signed distance field (SDF) and a textured field using two latent codes. DMTet is employed to derive a 3D surface mesh from the SDF, and we sample the texture field at the surface points to obtain color information. Our training incorporates adversarial losses focused on 2D images, specifically utilizing a rasterization-based differentiable renderer to produce both RGB images and silhouettes. To distinguish between genuine and generated inputs, we implement two separate 2D discriminators—one for RGB images and another for silhouettes. The entire framework is designed to be trainable in an end-to-end manner. As various sectors increasingly transition towards the development of expansive 3D virtual environments, the demand for scalable tools that can generate substantial quantities of high-quality and diverse 3D content has become apparent. Our research endeavors to create effective 3D generative models capable of producing textured meshes that can be seamlessly integrated into 3D rendering engines, thereby facilitating their immediate application in various downstream uses. This approach not only addresses the scalability challenge but also enhances the potential for innovative applications in virtual reality and gaming.
API Access
Has API
API Access
Has API
Integrations
No details available.
Integrations
No details available.
Pricing Details
Free
Free Trial
Free Version
Pricing Details
No price information available.
Free Trial
Free Version
Deployment
Web-Based
On-Premises
iPhone App
iPad App
Android App
Windows
Mac
Linux
Chromebook
Deployment
Web-Based
On-Premises
iPhone App
iPad App
Android App
Windows
Mac
Linux
Chromebook
Customer Support
Business Hours
Live Rep (24/7)
Online Support
Customer Support
Business Hours
Live Rep (24/7)
Online Support
Types of Training
Training Docs
Webinars
Live Training (Online)
In Person
Types of Training
Training Docs
Webinars
Live Training (Online)
In Person
Vendor Details
Company Name
DragGAN
Founded
2023
Website
vcai.mpi-inf.mpg.de/projects/DragGAN/
Vendor Details
Company Name
NVIDIA
Country
United States
Website
nv-tlabs.github.io/GET3D/