Equivariant Neural Rendering

Dupont Emilien
Dupont Emilien
Bautista Miguel Angel
Bautista Miguel Angel
Sankar Aditya
Sankar Aditya
Susskind Josh
Susskind Josh
Cited by: 0|Bibtex|Views119
Other Links: arxiv.org

Abstract:

We propose a framework for learning neural scene representations directly from images, without 3D supervision. Our key insight is that 3D structure can be imposed by ensuring that the learned representation transforms like a real 3D scene. Specifically, we introduce a loss which enforces equivariance of the scene representation with res...More

Code:

Data:

Your rating :
0

 

Tags
Comments