Vincent Lepetit
YOU?
Author Swipe
View article: Corr2Distrib: Making Ambiguous Correspondences an Ally to Predict Reliable 6D Pose Distributions
Corr2Distrib: Making Ambiguous Correspondences an Ally to Predict Reliable 6D Pose Distributions Open
We introduce Corr2Distrib, the first correspondence-based method which estimates a 6D camera pose distribution from an RGB image, explaining the observations. Indeed, symmetries and occlusions introduce visual ambiguities, leading to multi…
View article: Alligat0R: Pre-Training Through Co-Visibility Segmentation for Relative Camera Pose Regression
Alligat0R: Pre-Training Through Co-Visibility Segmentation for Relative Camera Pose Regression Open
Pre-training techniques have greatly advanced computer vision, with CroCo's cross-view completion approach yielding impressive results in tasks like 3D reconstruction and pose regression. However, cross-view completion is ill-posed in non-…
View article: NextBestPath: Efficient 3D Mapping of Unseen Environments
NextBestPath: Efficient 3D Mapping of Unseen Environments Open
This work addresses the problem of active 3D mapping, where an agent must find an efficient trajectory to exhaustively reconstruct a new scene. Previous approaches mainly predict the next best view near the agent's location, which is prone…
View article: UNIT: Unsupervised Online Instance Segmentation through Time
UNIT: Unsupervised Online Instance Segmentation through Time Open
Online object segmentation and tracking in Lidar point clouds enables autonomous agents to understand their surroundings and make safe decisions. Unfortunately, manual annotations for these tasks are prohibitively costly. We tackle this pr…
View article: BOP-Distrib: Revisiting 6D Pose Estimation Benchmarks for Better Evaluation under Visual Ambiguities
BOP-Distrib: Revisiting 6D Pose Estimation Benchmarks for Better Evaluation under Visual Ambiguities Open
6D pose estimation aims at determining the object pose that best explains the camera observation. The unique solution for non-ambiguous objects can turn into a multi-modal pose distribution for symmetrical objects or when occlusions of sym…
View article: NOPE: Novel Object Pose Estimation from a Single Image
NOPE: Novel Object Pose Estimation from a Single Image Open
The practicality of 3D object pose estimation remains limited for many applications due to the need for prior knowledge of a 3D model and a training period for new objects. To address this limitation, we propose an approach that takes a si…
View article: GuidedRec: Guiding Ill-Posed Unsupervised Volumetric Recovery
GuidedRec: Guiding Ill-Posed Unsupervised Volumetric Recovery Open
We introduce a novel unsupervised approach to reconstructing a 3D volume from only two planar projections that exploits a previous\-ly-captured 3D volume of the patient. Such volume is readily available in many important medical procedures…
View article: PyTorchGeoNodes: Enabling Differentiable Shape Programs for 3D Shape Reconstruction
PyTorchGeoNodes: Enabling Differentiable Shape Programs for 3D Shape Reconstruction Open
We propose PyTorchGeoNodes, a differentiable module for reconstructing 3D objects and their parameters from images using interpretable shape programs. Unlike traditional CAD model retrieval, shape programs allow reasoning about semantic pa…
View article: Gaussian Frosting: Editable Complex Radiance Fields with Real-Time Rendering
Gaussian Frosting: Editable Complex Radiance Fields with Real-Time Rendering Open
We propose Gaussian Frosting, a novel mesh-based representation for high-quality rendering and editing of complex 3D effects in real-time. Our approach builds on the recent 3D Gaussian Splatting framework, which optimizes a set of 3D Gauss…
View article: BEVContrast: Self-Supervision in BEV Space for Automotive Lidar Point Clouds
BEVContrast: Self-Supervision in BEV Space for Automotive Lidar Point Clouds Open
International audience
View article: BOP Challenge 2023 on Detection, Segmentation and Pose Estimation of Seen and Unseen Rigid Objects
BOP Challenge 2023 on Detection, Segmentation and Pose Estimation of Seen and Unseen Rigid Objects Open
We present the evaluation methodology, datasets and results of the BOP Challenge 2023, the fifth in a series of public competitions organized to capture the state of the art in model-based 6D object pose estimation from an RGB/RGB-D image …
View article: Correspondences of the Third Kind: Camera Pose Estimation from Object Reflection
Correspondences of the Third Kind: Camera Pose Estimation from Object Reflection Open
Computer vision has long relied on two kinds of correspondences: pixel correspondences in images and 3D correspondences on object surfaces. Is there another kind, and if there is, what can they do for us? In this paper, we introduce corres…
View article: GigaPose: Fast and Robust Novel Object Pose Estimation via One Correspondence
GigaPose: Fast and Robust Novel Object Pose Estimation via One Correspondence Open
We present GigaPose, a fast, robust, and accurate method for CAD-based novel object pose estimation in RGB images. GigaPose first leverages discriminative "templates", rendered images of the CAD models, to recover the out-of-plane rotation…
View article: SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering
SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering Open
We propose a method to allow precise and extremely fast mesh extraction from 3D Gaussian Splatting. Gaussian Splatting has recently become very popular as it yields realistic rendering while being significantly faster to train than NeRFs. …
View article: BEVContrast: Self-Supervision in BEV Space for Automotive Lidar Point Clouds
BEVContrast: Self-Supervision in BEV Space for Automotive Lidar Point Clouds Open
We present a surprisingly simple and efficient method for self-supervision of 3D backbone on automotive Lidar point clouds. We design a contrastive loss between features of Lidar scans captured in the same scene. Several such approaches ha…
View article: You Never Get a Second Chance To Make a Good First Impression: Seeding Active Learning for 3D Semantic Segmentation
You Never Get a Second Chance To Make a Good First Impression: Seeding Active Learning for 3D Semantic Segmentation Open
International audience
View article: HOC-Search: Efficient CAD Model and Pose Retrieval from RGB-D Scans
HOC-Search: Efficient CAD Model and Pose Retrieval from RGB-D Scans Open
We present an automated and efficient approach for retrieving high-quality CAD models of objects and their poses in a scene captured by a moving RGB-D camera. We first investigate various objective functions to measure similarity between a…
View article: CNOS: A Strong Baseline for CAD-based Novel Object Segmentation
CNOS: A Strong Baseline for CAD-based Novel Object Segmentation Open
We propose a simple three-stage approach to segment unseen objects in RGB images using their CAD models. Leveraging recent powerful foundation models, DINOv2 and Segment Anything, we create descriptors and generate proposals, including bin…
View article: In-Hand 3D Object Scanning from an RGB Sequence
In-Hand 3D Object Scanning from an RGB Sequence Open
International audience
View article: MACARONS: Mapping and Coverage Anticipation with RGB Online Self-Supervision
MACARONS: Mapping and Coverage Anticipation with RGB Online Self-Supervision Open
International audience
View article: Domain Transfer for 3D Pose Estimation from Color Images without Manual Annotations
Domain Transfer for 3D Pose Estimation from Color Images without Manual Annotations Open
We introduce a novel learning method for 3D pose estimation from color images. While acquiring annotations for color images is a difficult task, our approach circumvents this problem by learning a mapping from paired color and depth images…
View article: You Never Get a Second Chance To Make a Good First Impression: Seeding Active Learning for 3D Semantic Segmentation
You Never Get a Second Chance To Make a Good First Impression: Seeding Active Learning for 3D Semantic Segmentation Open
We propose SeedAL, a method to seed active learning for efficient annotation of 3D point clouds for semantic segmentation. Active Learning (AL) iteratively selects relevant data fractions to annotate within a given budget, but requires a f…
View article: NOPE: Novel Object Pose Estimation from a Single Image
NOPE: Novel Object Pose Estimation from a Single Image Open
The practicality of 3D object pose estimation remains limited for many applications due to the need for prior knowledge of a 3D model and a training period for new objects. To address this limitation, we propose an approach that takes a si…
View article: MACARONS: Mapping And Coverage Anticipation with RGB Online Self-Supervision
MACARONS: Mapping And Coverage Anticipation with RGB Online Self-Supervision Open
We introduce a method that simultaneously learns to explore new large environments and to reconstruct them in 3D from color images only. This is closely related to the Next Best View problem (NBV), where one has to identify where to move t…
View article: Deep Learning: Basics and Convolutional Neural Networks (CNNs)
Deep Learning: Basics and Convolutional Neural Networks (CNNs) Open
Deep learning belongs to the broader family of machine learning methods and currently provides state-of-the-art performance in a variety of fields, including medical applications. Deep learning architectures can be categorized into differe…
View article: Automatically Annotating Indoor Images with CAD Models via RGB-D Scans
Automatically Annotating Indoor Images with CAD Models via RGB-D Scans Open
International audience
View article: Back to MLP: A Simple Baseline for Human Motion Prediction
Back to MLP: A Simple Baseline for Human Motion Prediction Open
© 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, cre…
View article: A Simple and Powerful Global Optimization for Unsupervised Video Object Segmentation
A Simple and Powerful Global Optimization for Unsupervised Video Object Segmentation Open
International audience