Noah Snavely

70

Papers

258

Total Citations

Papers (70)

Generative Image Dynamics

LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias

Stereo4D: Learning How Things Move in 3D from Internet Stereo Videos

Beyond the Frame: Generating 360° Panoramic Videos from Perspective Videos

FlashDepth: Real-time Streaming Video Depth Estimation at 2K Resolution

Doppelgangers++: Improved Visual Disambiguation with Geometric 3D Features

Generating 3D-Consistent Videos from Unposed Internet Photos

WonderJourney: Going from Anywhere to Everywhere

Material Recognition in the Wild With the Materials in Context Database

DeepStereo: Learning to Predict New Views From the World's Imagery

Unsupervised Learning of Depth and Ego-Motion From Video

Shading Annotations in the Wild

Deep Feature Interpolation for Image Content Changes

MegaDepth: Learning Single-View Depth Prediction From Internet Photos

Learning Intrinsic Image Decomposition From Watching the World

Pushing the Boundaries of View Extrapolation With Multiplane Images

DeepView: View Synthesis With Learned Gradient Descent

Learning the Depths of Moving People by Watching Frozen People

Neural Rerendering in the Wild

TOUCHDOWN: Natural Language Navigation and Spatial Reasoning in Visual Street Environments

Depth Sensing Beyond LiDAR Range

Single-View View Synthesis With Multiplane Images

Lighthouse: Predicting Lighting Volumes for Spatially-Coherent Illumination

DualSDF: Semantic Shape Manipulation Using a Two-Level Representation

Visual Chirality

PhySG: Inverse Rendering With Spherical Gaussians for Physics-Based Material Editing and Relighting

Repopulating Street Scenes

KeypointDeformer: Unsupervised 3D Keypoint Discovery for Shape Control

Neural Scene Flow Fields for Space-Time View Synthesis of Dynamic Scenes

De-Rendering the World's Revolutionary Artefacts

Extreme Rotation Estimation Using Dense Correlation Volumes

Wide-Baseline Relative Camera Pose Estimation With Directional Learning

IBRNet: Learning Multi-View Image-Based Rendering

3D Moments From Near-Duplicate Photos

IRON: Inverse Rendering by Optimizing Neural SDFs and Materials From Photometric Images

Deformable Sprites for Unsupervised Video Decomposition

Persistent Nature: A Generative Model of Unbounded 3D Worlds

Seeing a Rose in Five Thousand Ways

Neural Scene Chronology

DynIBaR: Neural Dynamic Image-Based Rendering

Accidental Light Probes

Omnimatte3D: Associating Objects and Their Effects in Unconstrained Monocular Video

Neural Lens Modeling

BubbLeNet: Foveated Imaging for Visual Discovery

GeoStyle: Discovering Fashion Trends and Events

UprightNet: Geometry-Aware Camera Orientation Estimation From Single Images

Towers of Babel: Combining Images, Language, and 3D Geometry for Learning Multimodal Vision

Infinite Nature: Perpetual View Generation of Natural Scenes From a Single Image

Doppelgangers: Learning to Disambiguate Images of Similar Structures

Tracking Everything Everywhere All at Once

ASIC: Aligning Sparse in-the-wild Image Collections

Crowdsampling the Plenoptic Function

Learning Feature Descriptors using Camera Pose Supervision

Learning Gradient Fields for Shape Generation

Learning to Factorize and Relight a City

Hidden Footprints: Learning Contextual Walkability from 3D Human Trails

InfiniteNature-Zero: Learning Perpetual View Generation of Natural Scenes from Single Images

ARF: Artistic Radiance Fields

Structure and Motion from Casual Videos

Who's Waldo? Linking People Across Text and Images

Can Generative Video Models Help Pose Estimation?

MegaSaM: Accurate, Fast and Robust Structure and Motion from Casual Dynamic Videos

MoMaps: Semantics-Aware Scene Motion Generation with Motion Maps

Visual Chronicles: Using Multimodal LLMs to Analyze Massive Collections of Images

C3Po: Cross-View Cross-Modality Correspondence by Pointmap Prediction

NeRFiller: Completing Scenes via Generative 3D Inpainting

Discovery of Latent 3D Keypoints via End-to-end Geometric Reasoning

Multi-Plane Program Induction with 3D Box Priors

MetaSDF: Meta-Learning Signed Distance Functions

An Analysis of SVD for Deep Rotation Estimation