Methods and Systems for Rendering View-Dependent Images Using 2D Images

An image rendering system comprising:

• a processor;
• and
• memory with programming executable by the processor to:
• receive a plurality of 2D images of a 3D object;
• provide, to a neural network, the plurality of 2D images;
• generate a 3D neural model from the plurality of 2D images using a neural network;
• generate a triangular mesh using the 3D neural model;
• render a plurality of view-dependent images using the triangular mesh to generate a left view and a right view based on a head position of a viewer;
• and
• display, on a head mounted display, the left view to a left eye of the viewer and the right view to a right eye of the viewer.

An image rendering method for generating a plurality of view-dependent images at a display including a plurality of pixels, comprising:

• obtaining a 3D neural model from image data capturing a 3D shape of an object by obtaining a zero-level set of a signed distance function using a shape renderer of a rendering engine;
• modeling an appearance of the object by minimizing an image reconstruction error based upon the image data capturing the 3D shape of the object using an appearance renderer of the rendering engine;
• converting the neural model into a triangular mesh representing the object using the rendering engine;
• and
• rendering at least one image using the triangular mesh using the rendering engine.