Andrea Vedaldi

103

Papers

2,432

Total Citations

Papers (103)

Learning multiple visual domains with residual adapters

NeurIPS 2017arXiv

Learning feed-forward one-shot learners

NeurIPS 2016arXiv

Splatter Image: Ultra-Fast Single-View 3D Reconstruction

CoTracker3: Simpler and Better Point Tracking by Pseudo-Labelling Real Videos

Integrated perception with recurrent multi-task neural networks

NeurIPS 2016arXiv

GES : Generalized Exponential Splatting for Efficient Radiance Field Rendering

Free3D: Consistent Novel View Synthesis without 3D Representation

Learning the 3D Fauna of the Web

SHAP-EDITOR: Instruction-Guided Latent 3D Editing in Seconds

Unsupervised learning of object frames by dense equivariant image labelling

NeurIPS 2017arXiv

VMem: Consistent Interactive Video Scene Generation with Surfel-Indexed View Memory

AutoPartGen: Autoregressive 3D Part Generation and Discovery

UnCommon Objects in 3D

Tracktention: Leveraging Point Tracking to Attend Videos Faster and Better

Twinner: Shining Light on Digital Twins in a Few Snaps

Understanding Deep Image Representations by Inverting Them

Synthetic Data for Text Localisation in Natural Images

Weakly Supervised Deep Detection Networks

Dynamic Image Networks for Action Recognition

End-To-End Representation Learning for Correlation Filter Based Tracking

HPatches: A Benchmark and Evaluation of Handcrafted and Learned Local Descriptors

AnchorNet: A Weakly Supervised Network to Learn Geometry-Sensitive Features for Semantic Matching

Improved Texture Networks: Maximizing Quality and Diversity in Feed-Forward Stylization and Texture Synthesis

Self-Supervised Learning of Geometrically Stable Features Through Probabilistic Introspection

Efficient Parametrization of Multi-Domain Deep Neural Networks

MapNet: An Allocentric Spatial Memory for Mapping Environments

Net2Vec: Quantifying and Explaining How Concepts Are Encoded by Filters in Deep Neural Networks

Deep Image Prior

Slim DensePose: Thrifty Learning From Sparse Annotations and Motion Cues

Unsupervised Learning of Probably Symmetric Deformable 3D Objects From Images in the Wild

Transferring Dense Pose to Proximal Animal Classes

Self-Supervised Learning of Interpretable Keypoints From Unlabelled Videos

There and Back Again: Revisiting Backpropagation Saliency Methods

Localizing Visual Sounds the Hard Way

Pedestrian and Ego-Vehicle Trajectory Prediction From Monocular Camera

Unsupervised Learning of 3D Object Categories From Videos in the Wild

NeuroMorph: Unsupervised Shape Interpolation and Correspondence in One Go

Discovering Relationships Between Object Categories via Universal Canonical Maps

KeyTr: Keypoint Transporter for 3D Reconstruction of Deformable Objects in Videos

Generalized Category Discovery

BANMo: Building Animatable 3D Neural Models From Many Casual Videos

Self-Supervised Object Detection From Audio-Visual Correspondence

Deep Spectral Methods: A Surprisingly Strong Baseline for Unsupervised Semantic Segmentation and Localization

HOLODIFFUSION: Training a 3D Diffusion Model Using 2D Images

Novel-View Acoustic Synthesis

DynamicStereo: Consistent Dynamic Depth From Stereo Videos

RealFusion: 360deg Reconstruction of Any Object From a Single Image

Common Pets in 3D: Dynamic New-View Synthesis of Real-Life Deformable Categories

Continual Detection Transformer for Incremental Object Detection

MagicPony: Learning Articulated 3D Animals in the Wild

PC2: Projection-Conditioned Point Cloud Diffusion for Single-Image 3D Reconstruction

Interpretable Explanations of Black Boxes by Meaningful Perturbation

Learning 3D Object Categories by Looking Around Them

Unsupervised Learning of Object Landmarks by Factorized Spatial Embeddings

Understanding Deep Networks via Extremal Perturbations and Smooth Masks

Small Steps and Giant Leaps: Minimal Newton Solvers for Deep Learning

Unsupervised Learning of Landmarks by Descriptor Vector Exchange

C3DPO: Canonical 3D Pose Networks for Non-Rigid Structure From Motion

Learning to Discover Novel Visual Categories via Deep Transfer Clustering

Invariant Information Clustering for Unsupervised Image Classification and Segmentation

DensePose 3D: Lifting Canonical Surface Maps of Articulated Objects to the Third Dimension

On Compositions of Transformations in Contrastive Self-Supervised Learning

Space-Time Crop & Attend: Improving Cross-Modal Video Representation Learning

Replay: Multi-modal Multi-view Acted Videos for Casual Holography

Viewset Diffusion: (0-)Image-Conditioned 3D Generative Models from 2D Data

Online Clustered Codebook

What does CLIP know about a red circle? Visual prompt engineering for VLMs

HoloFusion: Towards Photo-realistic 3D Generative Modeling

End-to-End Visual Editing with a Generatively Pre-trained Artist

SNeS: Learning Probably Symmetric Neural Surfaces from Incomplete Data

Understanding Image Representations by Measuring Their Equivariance and Equivalence

DualPM: Dual Posed-Canonical Point Maps for 3D Shape and Pose Reconstruction

PartGen: Part-level 3D Generation and Reconstruction with Multi-view Diffusion Models

VGGT: Visual Geometry Grounded Transformer

Layered Motion Fusion: Lifting Motion Segmentation to 3D in Egocentric Videos

Dynamic Point Maps: A Versatile Representation for Dynamic 3D Reconstruction

Geo4D: Leveraging Video Generators for Geometric 4D Scene Reconstruction

SynCity: Training-Free Generation of 3D Cities

Back on Track: Bundle Adjustment for Dynamic Scene Reconstruction

DSO: Aligning 3D Generators with Simulation Feedback for Physical Soundness

Amodal3R: Amodal 3D Reconstruction from Occluded 2D Images

Puppet-Master: Scaling Interactive Video Generation as a Motion Prior for Part-Level Dynamics

IM-3D: Iterative Multiview Diffusion and Reconstruction for High-Quality 3D Generation

3D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexes

Deep Filter Banks for Texture Recognition and Segmentation

Modelling and unsupervised learning of symmetric deformable object categories

Unsupervised Learning of Object Landmarks through Conditional Image Generation

Gather-Excite: Exploiting Feature Context in Convolutional Neural Networks

Correlated Uncertainty for Learning Dense Correspondences from Noisy Labels

Fixing the train-test resolution discrepancy

Labelling unlabelled videos from scratch with multi-modal self-supervision

RELATE: Physically Plausible Multi-Object Scene Synthesis Using Structured Latent Spaces

Quantifying Learnability and Describability of Visual Concepts Emerging in Representation Learning

Continuous Surface Embeddings

3D Multi-bodies: Fitting Sets of Plausible 3D Human Models to Ambiguous Image Data

Canonical 3D Deformer Maps: Unifying parametric and non-parametric methods for dense weakly-supervised category reconstruction

Keeping Your Eye on the Ball: Trajectory Attention in Video Transformers

Unsupervised Part Discovery from Contrastive Reconstruction

Unsupervised Multi-Object Segmentation by Predicting Probable Motion Patterns

Contrastive Lift: 3D Object Instance Segmentation by Slow-Fast Contrastive Fusion

No Representation Rules Them All in Category Discovery

EPIC Fields: Marrying 3D Geometry and Video Understanding

Warped Convolutions: Efficient Invariance to Spatial Transformations