Posts by Tag

deep learning

3D Controllable GANs

We define the new task of 3D controllable image synthesis and propose an approach for solving it by reasoning both in 3D space and in the 2D image domain.

Differentiable Volumetric Rendering

Deep neural networks have revolutionized computer vision over the last decade. They excel in 2D-based vision tasks such as object detection, optical flow pre...

Texture Fields

Recently, deep learning methods in the 3D domain have gained popularity in the research community. One of the major goals in this area is to reconstruct 3D c...

Occupancy Flow

An intelligent agent that can interact with the world has to be able to reason in 3D. In recent year, there has therefore been a lot of interest in learning-...

Superquadrics Revisited

Recent advances in deep learning coupled with the abundance of large shape repositories gave rise to various methods that seek to learn the 3D model of an ob...

Occupancy Networks

In recent years, deep learning has led to many breakthroughs in computer vision. Many tasks such as object detection, semantic segmentation, optical flow est...

Back to Top ↑

3D reconstruction

Differentiable Volumetric Rendering

Deep neural networks have revolutionized computer vision over the last decade. They excel in 2D-based vision tasks such as object detection, optical flow pre...

Texture Fields

Recently, deep learning methods in the 3D domain have gained popularity in the research community. One of the major goals in this area is to reconstruct 3D c...

Occupancy Flow

An intelligent agent that can interact with the world has to be able to reason in 3D. In recent year, there has therefore been a lot of interest in learning-...

Superquadrics Revisited

Recent advances in deep learning coupled with the abundance of large shape repositories gave rise to various methods that seek to learn the 3D model of an ob...

Occupancy Networks

In recent years, deep learning has led to many breakthroughs in computer vision. Many tasks such as object detection, semantic segmentation, optical flow est...

Back to Top ↑

3D representations

Differentiable Volumetric Rendering

Deep neural networks have revolutionized computer vision over the last decade. They excel in 2D-based vision tasks such as object detection, optical flow pre...

Texture Fields

Recently, deep learning methods in the 3D domain have gained popularity in the research community. One of the major goals in this area is to reconstruct 3D c...

Occupancy Flow

An intelligent agent that can interact with the world has to be able to reason in 3D. In recent year, there has therefore been a lot of interest in learning-...

Superquadrics Revisited

Recent advances in deep learning coupled with the abundance of large shape repositories gave rise to various methods that seek to learn the 3D model of an ob...

Occupancy Networks

In recent years, deep learning has led to many breakthroughs in computer vision. Many tasks such as object detection, semantic segmentation, optical flow est...

Back to Top ↑

single view 3D reconstruction

Differentiable Volumetric Rendering

Deep neural networks have revolutionized computer vision over the last decade. They excel in 2D-based vision tasks such as object detection, optical flow pre...

Texture Fields

Recently, deep learning methods in the 3D domain have gained popularity in the research community. One of the major goals in this area is to reconstruct 3D c...

Occupancy Flow

An intelligent agent that can interact with the world has to be able to reason in 3D. In recent year, there has therefore been a lot of interest in learning-...

Occupancy Networks

In recent years, deep learning has led to many breakthroughs in computer vision. Many tasks such as object detection, semantic segmentation, optical flow est...

Back to Top ↑

autonomous vehicles

Learning to Drive in Cities

We propose a novel self-driving technique which addreses urban scenarios and doesn’t rely on detailed maps. This new approach learns high-level representatio...

Back to Top ↑

imitation learning

Learning to Drive in Cities

We propose a novel self-driving technique which addreses urban scenarios and doesn’t rely on detailed maps. This new approach learns high-level representatio...

Back to Top ↑

generative adversarial networks

3D Controllable GANs

We define the new task of 3D controllable image synthesis and propose an approach for solving it by reasoning both in 3D space and in the 2D image domain.

Back to Top ↑

primitive-based representations

Superquadrics Revisited

Recent advances in deep learning coupled with the abundance of large shape repositories gave rise to various methods that seek to learn the 3D model of an ob...

Back to Top ↑

self-driving

Learning to Drive in Cities

We propose a novel self-driving technique which addreses urban scenarios and doesn’t rely on detailed maps. This new approach learns high-level representatio...

Back to Top ↑

unsupervised

3D Controllable GANs

We define the new task of 3D controllable image synthesis and propose an approach for solving it by reasoning both in 3D space and in the 2D image domain.

Back to Top ↑

self driving

Back to Top ↑

unsupervised learning

Back to Top ↑

3d controllability

3D Controllable GANs

We define the new task of 3D controllable image synthesis and propose an approach for solving it by reasoning both in 3D space and in the 2D image domain.

Back to Top ↑

implicit neural representations

Back to Top ↑

depth fusion

Back to Top ↑

multi-view stereo

Back to Top ↑

affordances

Learning to Drive in Cities

We propose a novel self-driving technique which addreses urban scenarios and doesn’t rely on detailed maps. This new approach learns high-level representatio...

Back to Top ↑

4D reconstruction

Occupancy Flow

An intelligent agent that can interact with the world has to be able to reason in 3D. In recent year, there has therefore been a lot of interest in learning-...

Back to Top ↑

neural ordinary differential equations

Occupancy Flow

An intelligent agent that can interact with the world has to be able to reason in 3D. In recent year, there has therefore been a lot of interest in learning-...

Back to Top ↑

optical flow

Back to Top ↑

adversarial attacks

Back to Top ↑

motion deblur

Back to Top ↑

dagger

Back to Top ↑

structure-aware representations

Back to Top ↑

implicit neural representationss

Back to Top ↑

surface reconstruction

Back to Top ↑

point cloud

Back to Top ↑

semantic segmentation

Back to Top ↑

object detection

Back to Top ↑

3d awareness

Back to Top ↑

radiance fields

Back to Top ↑

causality

Back to Top ↑

counterfactuals

Back to Top ↑

generative models

Back to Top ↑

robustness

Back to Top ↑

image classification

Back to Top ↑

data augmentation

Back to Top ↑

multi-object tracking

Back to Top ↑

evaluation metrics

Back to Top ↑

visual tracking

Back to Top ↑

controllable image synthesis

Back to Top ↑

sensor fusion

Back to Top ↑

transformers

Back to Top ↑

stereo matching

Back to Top ↑

depth estimation

Back to Top ↑

stereo super-resolution

Back to Top ↑

neural coordinate-based representations

Back to Top ↑

real-time rendering

Back to Top ↑

novel view synthesis

Back to Top ↑

NeRF

Back to Top ↑

neural radiance fields

Back to Top ↑

KiloNeRF

Back to Top ↑

PlenOctrees

Back to Top ↑

Light Field Networks

Back to Top ↑

human avatar

Back to Top ↑

frequency bias

Back to Top ↑

frequency artifacts

Back to Top ↑

spectral bias

Back to Top ↑

deep fakes

Back to Top ↑

dataset

KITTI-360

With the recently released dataset KITTI-360, we develop a set of novel benchmarks to facilitate research at the intersection of vision, graphics and robotics.

Back to Top ↑