Massachusetts Institute of Technology

Publications

Controlling diverse robots by inferring Jacobian fields with deep networks

Nature • • Li • ✨Nature Journal Article

History-Guided Video Diffusion

ICLR • • Song, Chen

Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion

NeurIPS • • Chen
Thumbnail for publication "Neural Isometries: Taming Transformations for Equivariant ML"

Neural Isometries: Taming Transformations for Equivariant ML

NeurIPS • • Mitchel

FlowMap: High-Quality Camera Poses, Intrinsics, and Depth via Gradient Descent

3DV • • Smith, Charatan

pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction

CVPR • • Charatan • ✨Best Paper Runner-Up
Date
TitleSubtitle
First Authors Venue
Publication thumbnail

Locality in Image Diffusion Models Emerges from Data Statistics

Lukoianov NeurIPS Publication thumbnail

Meschers Geometry Processing of Impossible Objects

Dodik SIGGRAPH Publication thumbnail

Controlling diverse robots by inferring Jacobian fields with deep networks

Li ✨ Nature Publication thumbnail

History-Guided Video Diffusion

Song, Chen ICLR Publication thumbnail

Diffusion Forcing Next-token Prediction Meets Full-Sequence Diffusion

Chen NeurIPS Publication thumbnail

Neural Isometries Taming Transformations for Equivariant ML

Mitchel NeurIPS Publication thumbnail

Score Distillation via Reparametrized DDIM

Lukoianov NeurIPS Publication thumbnail

FlowMap High-Quality Camera Poses, Intrinsics, and Depth via Gradient Descent

Smith, Charatan 3DV Publication thumbnail

pixelSplat 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction

Charatan ✨ CVPR Publication thumbnail

Variational Barycentric Coordinates

Dodik ✨ SIGGRAPH Asia Publication thumbnail

Diffusion with Forward Models Solving Stochastic Inverse Problems Without Direct Supervision

Tewari, Yin ✨ NeurIPS Publication thumbnail

FlowCam Training Generalizable 3D Radiance Fields without Camera Poses via Pixel-Aligned Scene Flow

Smith NeurIPS Publication thumbnail

Learning to render novel views from wide-baseline stereo pairs

Du CVPR Publication thumbnail

Neural Groundplans Persistent Neural Scene Representations from a Single Image

Sharma ICLR Publication thumbnail

Decomposing NeRF for Editing via Feature Field Distillation

Kobayashi NeurIPS Publication thumbnail

Unsupervised Discovery and Composition of Object Light Fields

Smith TMLR Publication thumbnail

Neural Descriptor Fields SE(3)-Equivariant Object Representations for Manipulation

Simeonov, Du ICRA Publication thumbnail

Neural Fields in Visual Computing and Beyond

Xie Eurographics STAR Publication thumbnail

Learning Signal-Agnostic Manifolds of Neural Fields

Du NeurIPS Publication thumbnail

3D Neural Scene Representations for Visuomotor Control

Li, Li ✨ CoRL Publication thumbnail

Deep Medial Fields

Rebain arXiv Publication thumbnail

Light Field Networks Neural Scene Representations with Single-Evaluation Rendering

Sitzmann, Rezchikov ✨ NeurIPS Publication thumbnail

MetaSDF Meta-learning Signed Distance Functions

Sitzmann NeurIPS Publication thumbnail

Implicit Neural Representations with Periodic Activation Functions

Sitzmann, Martel ✨ NeurIPS Publication thumbnail

Scene Representation Networks Continuous 3D-Structure-Aware Neural Scene Representations

Sitzmann ✨ NeurIPS Publication thumbnail

DeepVoxels Learning Persistent 3D Feature Embeddings

Sitzmann ✨ CVPR
Thumbnail for publication "Locality in Image Diffusion Models Emerges from Data Statistics"

Locality in Image Diffusion Models Emerges from Data Statistics

NeurIPS • • Lukoianov

Meschers: Geometry Processing of Impossible Objects

SIGGRAPH • • Dodik

Controlling diverse robots by inferring Jacobian fields with deep networks

Nature • • Li • ✨Nature Journal Article

History-Guided Video Diffusion

ICLR • • Song, Chen

Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion

NeurIPS • • Chen
Thumbnail for publication "Neural Isometries: Taming Transformations for Equivariant ML"

Neural Isometries: Taming Transformations for Equivariant ML

NeurIPS • • Mitchel

Score Distillation via Reparametrized DDIM

NeurIPS • • Lukoianov

FlowMap: High-Quality Camera Poses, Intrinsics, and Depth via Gradient Descent

3DV • • Smith, Charatan

pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction

CVPR • • Charatan • ✨Best Paper Runner-Up
Thumbnail for publication "Variational Barycentric Coordinates"

Variational Barycentric Coordinates

SIGGRAPH Asia • • Dodik • ✨Journal Track

Diffusion with Forward Models: Solving Stochastic Inverse Problems Without Direct Supervision

NeurIPS • • Tewari, Yin • ✨Spotlight

FlowCam: Training Generalizable 3D Radiance Fields without Camera Poses via Pixel-Aligned Scene Flow

NeurIPS • • Smith

Learning to render novel views from wide-baseline stereo pairs

CVPR • • Du

Neural Groundplans: Persistent Neural Scene Representations from a Single Image

ICLR • • Sharma

Decomposing NeRF for Editing via Feature Field Distillation

NeurIPS • • Kobayashi
Thumbnail for publication "Unsupervised Discovery and Composition of Object Light Fields"

Unsupervised Discovery and Composition of Object Light Fields

TMLR • • Smith

Neural Descriptor Fields: SE(3)-Equivariant Object Representations for Manipulation

ICRA • • Simeonov, Du
Thumbnail for publication "Neural Fields in Visual Computing and Beyond"

Neural Fields in Visual Computing and Beyond

Eurographics STAR • • Xie

Learning Signal-Agnostic Manifolds of Neural Fields

NeurIPS • • Du
Thumbnail for publication "3D Neural Scene Representations for Visuomotor Control"

3D Neural Scene Representations for Visuomotor Control

CoRL • • Li, Li • ✨Oral
Thumbnail for publication "Deep Medial Fields"

Deep Medial Fields

arXiv • • Rebain

Light Field Networks: Neural Scene Representations with Single-Evaluation Rendering

NeurIPS • • Sitzmann, Rezchikov • ✨Spotlight

MetaSDF: Meta-learning Signed Distance Functions

NeurIPS • • Sitzmann

Implicit Neural Representations with Periodic Activation Functions

NeurIPS • • Sitzmann, Martel • ✨Oral

Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representations

NeurIPS • • Sitzmann • ✨Outstanding New Directions - Honorable Mention

DeepVoxels: Learning Persistent 3D Feature Embeddings

CVPR • • Sitzmann • ✨Oral