ECCV 2024 accepted papers on 3D vision
548 of the 2,387 papers accepted at ECCV 2024 (23.0%) are on 3D vision, compared with 18.6% at ECCV 2022. They fall into 6 topics.
See ECCV 2024 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
3D generation · 142 papers
- LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation (494 citations)
- SV3D: Novel Multi-view Synthesis and 3D Generation from a Single Image using Latent Video Diffusion (262 citations)
- GeoWizard: Unleashing the Diffusion Priors for 3D Geometry Estimation from a Single Image (156 citations)
- OccWorld: Learning a 3D Occupancy World Model for Autonomous Driving (156 citations)
- GaussianFormer: Scene as Gaussians for Vision-Based 3D Semantic Occupancy Prediction (108 citations)
- latentSplat: Autoencoding Variational Gaussians for Fast Generalizable 3D Reconstruction (90 citations)
- SceneVerse: Scaling 3D Vision-Language Learning for Grounded Scene Understanding (90 citations)
- Controllable Human-Object Interaction Synthesis (86 citations)
- Fully Sparse 3D Occupancy Prediction (81 citations)
- DreamScene: 3D Gaussian-based Text-to-3D Scene Generation via Formation Pattern Sampling (60 citations)
- ReMoS: 3D Motion-Conditioned Reaction Synthesis for Two-Person Interactions (57 citations)
- Generating Human Interaction Motions in Scenes with Text Control (52 citations)
and 130 more in the interactive map.
Gaussian splatting · 96 papers
- MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Images (409 citations)
- Gaussian Grouping: Segment and Edit Anything in 3D Scenes (338 citations)
- FSGS: Real-Time Few-shot View Synthesis using Gaussian Splatting (309 citations)
- Street Gaussians: Modeling Dynamic Urban Scenes with Gaussian Splatting (293 citations)
- GRM: Large Gaussian Reconstruction Model for Efficient 3D Reconstruction and Generation (252 citations)
- Relightable 3D Gaussians: Realistic Point Cloud Relighting with BRDF Decomposition and Ray Tracing (247 citations)
- CityGaussian: Real-time High-quality Large-Scale Scene Rendering with Gaussians (235 citations)
- Mini-Splatting: Representing Scenes with a Constrained Number of Gaussians (224 citations)
- GS-LRM: Large Reconstruction Model for 3D Gaussian Splatting (209 citations)
- CompGS: Smaller and Faster Gaussian Splatting with Vector Quantization (184 citations)
- EAGLES: Efficient Accelerated 3D Gaussians with Lightweight EncodingS (183 citations)
- HAC: Hash-grid Assisted Context for 3D Gaussian Splatting Compression (167 citations)
and 84 more in the interactive map.
Face and shape · 85 papers
- Grounding Image Matching in 3D with MASt3R (563 citations)
- CRM: Single Image to 3D Textured Mesh with Convolutional Reconstruction Model (193 citations)
- Expressive Whole-Body 3D Gaussian Avatar (74 citations)
- TalkingGaussian: Structure-Persistent 3D Talking Head Synthesis via Gaussian Splatting (61 citations)
- FlashTex: Fast Relightable Mesh Texturing with LightControlNet (42 citations)
- Disentangled Clothed Avatar Generation from Text Descriptions (38 citations)
- MeshAvatar: Learning High-quality Triangular Human Avatars from Multi-view Videos (35 citations)
- ScanTalk: 3D Talking Heads from Unregistered Scans (25 citations)
- NICP: Neural ICP for 3D Human Registration at Scale (21 citations)
- GarmentCodeData: A Dataset of 3D Made-to-Measure Garments With Sewing Patterns (18 citations)
- iHuman: Instant Animatable Digital Humans From Monocular Videos (18 citations)
- EMIE-MAP: Large-Scale Road Surface Reconstruction Based on Explicit Mesh and Implicit Encoding (16 citations)
and 73 more in the interactive map.
Human pose estimation · 85 papers
- Sapiens: Foundation for Human Vision Models (218 citations)
- Champ: Controllable and Consistent Human Image Animation with 3D Parametric Guidance (177 citations)
- FoundPose: Unseen Object Pose Estimation with Foundation Features (88 citations)
- TRAM: Global Trajectory and Motion of 3D Humans from in-the-wild Videos (72 citations)
- Multi-HMR: Multi-Person Whole-Body Human Mesh Recovery in a Single Shot (71 citations)
- FreeZe: Training-free zero-shot 6D pose estimation with geometric and vision foundation models (48 citations)
- EgoPoser: Robust Real-Time Egocentric Pose Estimation from Sparse and Intermittent Observations Everywhere (41 citations)
- Omni6DPose: A Benchmark and Model for Universal 6D Object Pose Estimation and Tracking (34 citations)
- A Graph-Based Approach for Category-Agnostic Pose Estimation (31 citations)
- WorldPose: A World Cup Dataset for Global 3D Human Pose Estimation (30 citations)
- SpaRP: Fast 3D Object Reconstruction and Pose Estimation from Sparse Views (27 citations)
- CoMusion: Towards Consistent Stochastic Human Motion Prediction via Motion Diffusion (26 citations)
and 73 more in the interactive map.
Depth and stereo · 82 papers
- Global Structure-from-Motion Revisited (162 citations)
- DiffusionDepth: Diffusion Denoising Approach for Monocular Depth Estimation (125 citations)
- MapTracker: Tracking with Strided Memory Fusion for Consistent Vector HD Mapping (54 citations)
- Scene Coordinate Reconstruction: Posing of Image Collections via Incremental Learning of a Relocalizer (52 citations)
- Cross-view image geo-localization with Panorama-BEV Co-Retrieval Network (46 citations)
- ConGeo: Robust Cross-view Geo-localization across Ground View Variations (39 citations)
- GeoCalib: Learning Single-image Calibration with Geometric Optimization (39 citations)
- Diffusion Models for Monocular Depth Estimation: Overcoming Challenging Conditions (34 citations)
- Six-Point Method for Multi-Camera Systems with Reduced Solution Space (30 citations)
- Motion-prior Contrast Maximization for Dense Continuous-Time Motion Estimation (27 citations)
- OGNI-DC: Robust Depth Completion with Optimization-Guided Neural Iterations (26 citations)
- Mono-ViFI: A Unified Learning Framework for Self-supervised Single- and Multi-frame Monocular Depth Estimation (23 citations)
and 70 more in the interactive map.
NeRF and radiance fields · 58 papers
- Generative Camera Dolly: Extreme Monocular Dynamic Novel View Synthesis (51 citations)
- LaRa: Efficient Large-Baseline Radiance Fields (45 citations)
- FisherRF: Active View Selection and Mapping with Radiance Fields using Fisher Information (44 citations)
- The Nerfect Match: Exploring NeRF Features for Visual Localization (31 citations)
- Protecting NeRFs' Copyright via Plug-And-Play Watermarking Base Model (21 citations)
- NeRF-MAE: Masked AutoEncoders for Self-Supervised 3D Representation Learning for Neural Radiance Fields (20 citations)
- SphereHead: Stable 3D Full-head Synthesis with Spherical Tri-plane Representation (20 citations)
- BeNeRF:Neural Radiance Fields from a Single Blurry Image and Event Stream (15 citations)
- KFD-NeRF: Rethinking Dynamic NeRF with Kalman Filter (15 citations)
- Taming Latent Diffusion Model for Neural Radiance Field Inpainting (15 citations)
- City-on-Web: Real-time Neural Rendering of Large-scale Scenes on the Web (14 citations)
- Leveraging Thermal Modality to Enhance Reconstruction in Low-Light Conditions (14 citations)
and 46 more in the interactive map.
