3d-from-multi-view-and-sensors.md

CVPR-2023-Papers

Application
New collections

3D from Multi-View and Sensors

Title	Repo	Paper	Video
NeuMap: Neural Coordinate Mapping by Auto-Transdecoder for Camera Localization
Object Pose Estimation with Statistical Guarantees: Conformal Keypoint Detection and Geometric Uncertainty Propagation
NeuralUDF: Learning Unsigned Distance Fields for Multi-View Reconstruction of Surfaces with Arbitrary Topologies
NEF: Neural Edge Fields for 3D Parametric Curve Reconstruction from Multi-View Images
Looking through the Glass: Neural Surface Reconstruction Against High Specular Reflections
Multi-View Azimuth Stereo via Tangent Space Consistency			➖
Instant Multi-View Head Capture through Learnable Registration
EditableNeRF: Editing Topologically Varying Neural Radiance Fields by Key Points
Iterative Geometry Encoding Volume for Stereo Matching			➖
Hi-LASSIE: High-Fidelity Articulated Shape and Skeleton Discovery from Sparse Image Ensemble
VDN-NeRF: Resolving Shape-Radiance Ambiguity via View-Dependence Normalization
Neuralangelo: High-Fidelity Neural Surface Reconstruction
In-Hand 3D Object Scanning from an RGB Sequence
SHS-Net: Learning Signed Hyper Surfaces for Oriented Normal Estimation of Point Clouds
FAC: 3D Representation Learning via Foreground Aware Feature Contrast			➖
Neural Kernel Surface Reconstruction
NeRFVS: Neural Radiance Fields for Free View Synthesis via Geometry Scaffolds	➖		➖
HelixSurf: A Robust and Efficient Neural Implicit Surface Learning of Indoor Scenes with Iterative Intertwined Regularization
Multi-Space Neural Radiance Fields
MSF: Motion-guided Sequential Fusion for Efficient 3D Object Detection from Point Cloud Sequences			➖
PVO: Panoptic Visual Odometry
Diffusion-SDF: Text-to-Shape via Voxelized Diffusion
Rotation-Invariant Transformer for Point Cloud Matching
HexPlane: A Fast Representation for Dynamic Scenes
Learning 3D Representations from 2D Pre-trained Models via Image-to-Point Masked Autoencoders			➖
Progressive Neighbor Consistency Mining for Correspondence Pruning
SCoDA: Domain Adaptive Shape Completion for Real Scans			➖
Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo
Level-S2fM: Structure from Motion on Neural Level Set of Implicit Surfaces
PLA: Language-Driven Open-Vocabulary 3D Scene Understanding
SUDS: Scalable Urban Dynamic Scenes
3D Semantic Segmentation in the Wild: Learning Generalized Models for Adverse-Condition Point Clouds
BAEFormer: Bi-Directional and Early Interaction Transformers for Bird's Eye View Semantic Segmentation	➖		➖
Dionysus: Recovering Scene Structures by Dividing into Semantic Pieces	➖		➖
LP-DIF: Learning Local Pattern-Specific Deep Implicit Function for 3D Objects and Scenes	➖		➖
Neural Kaleidoscopic Space Sculpting
Starting from Non-Parametric Networks for 3D Point Cloud Analysis
Panoptic Compositional Feature Field for Editable Scene Rendering with Network-Inferred Labels via Metric Learning	➖		➖
Robust Dynamic Radiance Fields
BAD-NeRF: Bundle Adjusted Deblur Neural Radiance Fields
Consistent Direct Time-of-Flight Video Depth Super-Resolution
Patch-based 3D Natural Scene Generation from a Single Example
3D Video Loops from Asynchronous Input
UniDistill: A Universal Cross-Modality Knowledge Distillation Framework for 3D Object Detection in Bird's-Eye View
Neural Scene Chronology
RUST: Latent Neural Scene Representations from Unposed Imagery			➖
Painting 3D Nature in 2D: View Synthesis of Natural Scenes from a Single Semantic Mask
F2-NeRF: Fast Neural Radiance Field Training with Free Camera Trajectories
VL-SAT: Visual-Linguistic Semantics Assisted Training for 3D Semantic Scene Graph Prediction in Point Cloud
REC-MV: REconstructing 3D Dynamic Cloth from Monocular Videos
MVImgNet: A Large-Scale Dataset of Multi-View Images			➖
Shakes on a Plane: Unsupervised Depth Estimation from Unstabilized Photography
GINA-3D: Learning to Generate Implicit Neural Assets in the Wild	➖
MobileNeRF: Exploiting the Polygon Rasterization Pipeline for Efficient Neural Field Rendering on Mobile Architectures
DynIBaR: Neural Dynamic Image-based Rendering
IMP: Iterative Matching and Pose Estimation with Adaptive Pooling
Learning the Distribution of Errors in Stereo Matching for Joint Disparity and Uncertainty Estimation
NeAT: Learning Neural Implicit Surfaces with Arbitrary Topologies from Multi-View Images
ShadowNeuS: Neural SDF Reconstruction by Shadow Ray Supervision
Bi-LRFusion: Bi-Directional LiDAR-Radar Fusion for 3D Dynamic Object Detection
NeRF-DS: Neural Radiance Fields for Dynamic Specular Objects
LoGoNet: Towards Accurate 3D Object Detection with Local-to-Global Cross-Modal Fusion			➖
3D Registration with Maximal Cliques
OmniObject3D: Large-Vocabulary 3D Object Dataset for Realistic Perception, Reconstruction and Generation
Progressive Spatio-Temporal Alignment for Efficient Event-based Motion Estimation
RefSR-NeRF: Towards High Fidelity and Super Resolution View Synthesis	➖		➖
NoPe-NeRF: Optimising Neural Radiance Field with no Pose Prior
Spherical Transformer for LiDAR-based 3D Recognition
Progressively Optimized Local Radiance Fields for Robust View Synthesis
PersonNeRF: Personalized Reconstruction from Photo Collections
NeuWigs: A Neural Dynamic Model for Volumetric Hair Capture and Animation
Representing Volumetric Videos as Dynamic MLP Maps
Rethinking the Approximation Error in 3D Surface Fitting for Point Cloud Normal Estimation
A Practical Stereo Depth System for Smart Glasses	➖
Compressing Volumetric Radiance Fields to 1 MB			➖
HyperReel: High-Fidelity 6-DoF Video with Ray-Conditioned Sampling
Point2Pix: Photo-Realistic Point Cloud Rendering via Neural Radiance Fields	➖
Command-Driven Articulated Object Understanding and Manipulation	➖
SCADE: NeRFs from Space Carving with Ambiguity-Aware Depth Estimates
Panoptic Lifting for 3D Scene Understanding with Neural Fields
PaletteNeRF: Palette-based Appearance Editing of Neural Radiance Fields			➖
NeRFLix: High-Quality Neural View Synthesis by Learning a Degradation-Driven Inter-Viewpoint MiXer			➖
SegLoc: Learning Segmentation-based Representations for Privacy-Preserving Visual Localization	➖		➖
expOSE: Accurate Initialization-Free Projective Factorization using Exponential Regularization	➖		➖
Neural Vector Fields: Implicit Representation by Explicit Learning
Unsupervised Inference of Signed Distance Functions from Single Sparse Point Clouds without Learning Priors
Learning to Measure the Point Cloud Reconstruction Loss in a Representation Space	➖		➖
Grad-PU: Arbitrary-Scale Point Cloud Upsampling via Gradient Descent with Learned Distance Functions
TensoIR: Tensorial Inverse Rendering
Multi-View Inverse Rendering for Large-Scale Real-World Indoor Scenes
Frequency-Modulated Point Cloud Rendering with Easy Editing
VoxelNeXt: Fully Sparse VoxelNet for 3D Object Detection and Tracking
RGBD2: Generative Scene Synthesis via Incremental View Inpainting using RGBD Diffusion Models
Multi-View Stereo Representation Revist: Region-Aware MVSNet	➖
AutoRecon: Automated 3D Object Discovery and Reconstruction
Common Pets in 3D: Dynamic New-View Synthesis of Real-Life Deformable Categories	➖		➖
Binarizing Sparse Convolutional Networks for Efficient Point Cloud Analysis	➖		➖
LargeKernel3D: Scaling Up Kernels in 3D Sparse CNNs
Learning 3D Scene Priors With 2D Supervision
NeuralEditor: Editing Neural Radiance Fields via Manipulating Point Clouds
NeuralPCI: Spatio-Temporal Neural Field for 3D Point Cloud Multi-Frame Non-Linear Interpolation
Two-View Geometry Scoring Without Correspondences
Deep Graph-Based Spatial Consistency for Robust Non-Rigid Point Cloud Registration
RIAV-MVS: Recurrent-Indexing an Asymmetric Volume for Multi-View Stereo	➖
Depth Estimation From Camera Image and mmWave Radar Point Cloud			➖
Hierarchical Supervision and Shuffle Data Augmentation for 3D Semi-Supervised Object Detection
Multiscale Tensor Decomposition and Rendering Equation Encoding for View Synthesis
PATS: Patch Area Transportation With Subdivision for Local Feature Matching
Depth Estimation From Indoor Panoramas With Neural Scene Representation
Masked Representation Learning for Domain Generalized Stereo Matching
GANHead: Towards Generative Animatable Neural Head Avatars
Visual-Tactile Sensing for In-Hand Object Reconstruction
IterativePFN: True Iterative Point Cloud Filtering
Towards Better Gradient Consistency for Neural Signed Distance Functions via Level Set Alignment			➖
GarmentTracking: Category-Level Garment Pose Tracking
Learning Transformation-Predictive Representations for Detection and Description of Local Features	➖		➖
Local Implicit Ray Function for Generalizable Radiance Field Representation
Grid-Guided Neural Radiance Fields for Large Urban Scenes
EventNeRF: Neural Radiance Fields From a Single Colour Event Camera
Learning Optical Expansion From Scale Matching
Self-Supervised 3D Scene Flow Estimation Guided by Superpoints
Adaptive Annealing for Robust Geometric Estimation	➖
SurfelNeRF: Neural Surfel Radiance Fields for Online Photorealistic Reconstruction of Indoor Scenes
PlaneDepth: Self-Supervised Depth Estimation via Orthogonal Planes
High-Res Facial Appearance Capture From Polarized Smartphone Images
Tensor4D: Efficient Neural 4D Decomposition for High-Fidelity Dynamic Reconstruction and Rendering			➖
Fully Self-Supervised Depth Estimation From Defocus Clue			➖
Adaptive Assignment for Geometry Aware Local Feature Matching
Efficient Second-Order Plane Adjustment	➖
Learning Adaptive Dense Event Stereo From the Image Domain	➖
FreeNeRF: Improving Few-Shot Neural Rendering With Free Frequency Regularization
SteerNeRF: Accelerating NeRF Rendering via Smooth Viewpoint Trajectory
Cross-Guided Optimization of Radiance Fields With Multi-View Image Super-Resolution for High-Resolution Novel View Synthesis	➖
AeDet: Azimuth-Invariant Multi-View 3D Object Detection
Local-to-Global Registration for Bundle-Adjusting Neural Radiance Fields
DKM: Dense Kernelized Feature Matching for Geometry Estimation
DINER: Depth-Aware Image-Based NEural Radiance Fields
HGNet: Learning Hierarchical Geometry From Points, Edges, and Surfaces	➖
Instant Volumetric Head Avatars
3D Line Mapping Revisited
Learning To Fuse Monocular and Multi-View Cues for Multi-Frame Depth Estimation in Dynamic Scenes
ESLAM: Efficient Dense SLAM System Based on Hybrid Representation of Signed Distance Fields
Point Cloud Forecasting as a Proxy for 4D Occupancy Forecasting
SparsePose: Sparse-View Camera Pose Regression and Refinement	➖		➖
Controllable Mesh Generation Through Sparse Latent Point Diffusion Models			➖
ARO-Net: Learning Implicit Fields From Anchored Radial Observations
Semantic Ray: Learning a Generalizable Semantic Field With Cross-Reprojection Attention
Sphere-Guided Training of Neural Implicit Surfaces
Finding Geometric Models by Clustering in the Consensus Space
NeuralDome: A Neural Modeling Pipeline on Multi-View Human-Object Interactions
Privacy-Preserving Representations Are Not Enough: Recovering Scene Content From Camera Poses
Robust Multiview Point Cloud Registration With Reliable Pose Graph Initialization and History Reweighting
Neural Part Priors: Learning To Optimize Part-Based Object Completion in RGB-D Scans
Accelerated Coordinate Encoding: Learning to Relocalize in Minutes Using RGB and Poses
Gated Stereo: Joint Depth Estimation From Gated and Wide-Baseline Active Stereo Cues
Revisiting Rotation Averaging: Uncertainties and Robust Losses			➖
NeRF-Supervised Deep Stereo
POEM: Reconstructing Hand in a Point Embedded Multi-View Stereo
vMAP: Vectorised Object Mapping for Neural Field SLAM
PET-NeuS: Positional Encoding Tri-Planes for Neural Surfaces
Learnable Skeleton-Aware 3D Point Cloud Sampling	➖
ObjectMatch: Robust Registration Using Canonical Object Correspondences
DiffRF: Rendering-Guided 3D Radiance Field Diffusion
Learning a Depth Covariance Function			➖
Viewpoint Equivariance for Multi-View 3D Object Detection
BlendFields: Few-Shot Example-Driven Facial Modeling
Implicit Surface Contrastive Clustering for LiDAR Point Clouds	➖
Self-Supervised Super-Plane for Neural 3D Reconstruction	➖
DiffusioNeRF: Regularizing Neural Radiance Fields with Denoising Diffusion Models
AligNeRF: High-Fidelity Neural Radiance Fields via Alignment-Aware Training			➖
VisFusion: Visibility-Aware Online 3D Scene Reconstruction from Videos
Fast Monocular Scene Reconstruction with Global-Sparse Local-Dense Grids
Semi-Weakly Supervised Object Kinematic Motion Prediction
OmniVidar: Omnidirectional Depth Estimation from Multi-Fisheye Images	➖
ContraNeRF: Generalizable Neural Radiance Fields for Synthetic-to-Real Novel View Synthesis via Contrastive Learning	➖		➖
PointVector: A Vector Representation in Point Cloud Analysis	➖		➖
Poly-PC: A Polyhedral Network for Multiple Point Cloud Tasks at Once	➖
Learning Neural Duplex Radiance Fields for Real-Time View Synthesis
VolRecon: Volume Rendering of Signed Ray Distance Functions for Generalizable Multi-View Reconstruction
CompletionFormer: Depth Completion with Convolutions and Vision Transformers
Exact-NeRF: An Exploration of a Precise Volumetric Parameterization for Neural Radiance Fields
Collaboration Helps Camera Overtake LiDAR in 3D Detection
SPIn-NeRF: Multiview Segmentation and Perceptual Inpainting with Neural Radiance Fields
GeoMVSNet: Learning Multi-View Stereo with Geometry Perception
3D Shape Reconstruction of Semi-Transparent Worms	➖
Revisiting Rolling Shutter Bundle Adjustment: Toward Accurate and Fast Solution
Virtual Occlusions through Implicit Depth
Neural Fields meet Explicit Geometric Representations for Inverse Rendering of Urban Scenes
Building Rearticulable Models for Arbitrary 3D Objects from 4D Point Clouds
DynamicStereo: Consistent Dynamic Depth from Stereo Videos			➖
Robust Outlier Rejection for 3D Registration with Variational Bayes			➖
Meta Architecture for Point Cloud Analysis
DyLiN: Making Light Field Networks Dynamic
Domain Generalized Stereo Matching via Hierarchical Visual Transformation
Learning Visibility Field for Detailed 3D Human Reconstruction and Relighting	➖
LightedDepth: Video Depth Estimation in Light of Limited Inference View Angles			➖
Long-Term Visual Localization with Mobile Sensors
Revisiting the P3P Problem	➖
I2-SDF: Intrinsic Indoor Scene Reconstruction and Editing via Raytracing in Neural SDFs
WildLight: In-the-Wild Inverse Rendering with a Flashlight
SE-ORNet: Self-Ensembling Orientation-Aware Network for Unsupervised Point Cloud Shape Correspondence
Teleidoscopic Imaging System for Microscale 3D Shape Reconstruction	➖
NeuDA: Neural Deformable Anchor for High-Fidelity Implicit Surface Reconstruction			➖
PointClustering: Unsupervised Point Cloud Pre-Training using Transformation Invariance in Clustering
PermutoSDF: Fast Multi-View Reconstruction with Implicit Surfaces using Permutohedral Lattices
TriVol: Point Cloud Rendering via Triple Volumes
Towards Unbiased Volume Rendering of Neural Implicit Surfaces with Geometry Priors	➖
Semi-Supervised Stereo-based 3D Object Detection via Cross-View Consensus	➖		➖
Self-Supervised Pre-Training with Masked Shape Prediction for 3D Scene Understanding	➖
Octree Guided Unoriented Surface Reconstruction
Towards Domain Generalization for Multi-View 3D Object Detection in Bird-Eye-View	➖
Learning Neural Volumetric Representations of Dynamic Humans in Minutes
AnchorFormer: Point Cloud Completion from Discriminative Nodes
Transforming Radiance Field with Lipschitz Network for Photorealistic 3D Scene Stylization	➖		➖
GANmouflage: 3D Object Nondetection with Texture Fields			➖
PEAL: Prior-embedded Explicit Attention Learning for Low-Overlap Point Cloud Registration
NeRFLight: Fast and Light Neural Radiance Fields using a Shared Feature Grid	➖
TMO: Textured Mesh Acquisition of Objects with a Mobile Device by using Differentiable Rendering
Generating Part-Aware Editable 3D Shapes without 3D Supervision
ALTO: Alternating Latent Topologies for Implicit 3D Reconstruction			➖
ORCa: Glossy Objects as Radiance-Field Cameras			➖
NeFII: Inverse Rendering for Reflectance Decomposition with Near-Field Indirect Illumination
BEV-guided Multi-Modality Fusion for Driving Perception			➖
K-Planes: Explicit Radiance Fields in Space, Time, and Appearance
RobustNeRF: Ignoring Distractors with Robust Losses
Unsupervised Deep Asymmetric Stereo Matching with Spatially-Adaptive Self-Similarity	➖		➖
ProxyFormer: Proxy Alignment Assisted Point Cloud Completion with Missing Part Sensitive Transformer			➖
Diffusion-based Signed Distance Fields for 3D Shape Generation
FeatureBooster: Boosting Feature Descriptors with a Lightweight Neural Network
Temporal Interpolation is All You Need for Dynamic Neural Radiance Fields
Neural Lens Modeling
Multi-View Reconstruction using Signed Ray Distance Functions (SRDF)	➖		➖
Masked Wavelet Representation for Compact Neural Radiance Fields
A Rotation-Translation-Decoupled Solution for Robust and Efficient Visual-Inertial Initialization
MACARONS: Mapping and Coverage Anticipation with RGB Online Self-Supervision
DualRefine: Self-Supervised Depth and Pose Estimation through Iterative Epipolar Sampling and Refinement Toward Equilibrium
CLIP2: Contrastive Language-Image-Point Pretraining from Real-World Point Cloud Data
Semidefinite Relaxations for Robust Multiview Triangulation
High-Frequency Stereo Matching Network
CAP: Robust Point Cloud Classification via Semantic and Structural Modeling	➖
Co-SLAM: Joint Coordinate and Sparse Parametric Encodings for Neural Real-Time SLAM
Temporally Consistent Online Depth Estimation using Point-based Fusion
Learning Neural Parametric Head Models
PointConvFormer: Revenge of the Point-based Convolution
Four-View Geometry with Unknown Radial Distortion
Seeing through the Glass: Neural 3D Reconstruction of Object Inside a Transparent Container