From the lab

ShinyNeRF: Modeling Anisotropic Reflections in Neural Radiance Fields

A physically-based NeRF variant that jointly estimates geometry and anisotropic specular parameters from multi-view images.

• Neural Radiance Fields
• view synthesis
• specular reflections

Read article →

Listen and Move: Improving GANs Coherency in Agnostic Sound-to-Video Generation

A cross-modal adversarial network equipped with dilated, multiscale RNN and video prediction layers for generic sound-to-video generation.

• generative artificial intelligence
• generative adversarial networks
• cross-modal generation

Read article →

Extended Labeled Faces in-the-Wild (ELFW): Augmenting Classes for Face Segmentation

An extended segmentation dataset with face-related occluding objects.

• semantic segmentation
• deeplabv3
• fully convolutional networks

Read article →