CVPR 2024 accepted papers on 3D vision
631 of the 2,716 papers accepted at CVPR 2024 (23.2%) are on 3D vision, compared with 20.8% at CVPR 2023. They fall into 6 topics.
See CVPR 2024 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
3D generation · 169 papers
- Wonder3D: Single Image to 3D using Cross-Domain Diffusion (341 citations)
- GaussianDreamer: Fast Generation from Text to 3D Gaussians by Bridging 2D and 3D Diffusion Models (141 citations)
- One-2-3-45++: Fast Single Image to 3D Objects with Consistent Multi-View Generation and 3D Diffusion (141 citations)
- LucidDreamer: Towards High-Fidelity Text-to-3D Generation via Interval Score Matching (118 citations)
- DiffuScene: Denoising Diffusion Models for Generative Indoor Scene Synthesis (93 citations)
- Holodeck: Language Guided Generation of 3D Embodied AI Environments (91 citations)
- Language Embedded 3D Gaussians for Open-Vocabulary Scene Understanding (80 citations)
- EMAGE: Towards Unified Holistic Co-Speech Gesture Generation via Expressive Masked Audio Gesture Modeling (79 citations)
- Editable Scene Simulation for Autonomous Driving via Collaborative LLM-Agents (73 citations)
- XCube: Large-Scale 3D Generative Modeling using Sparse Voxel Hierarchies (69 citations)
- RichDreamer: A Generalizable Normal-Depth Diffusion Model for Detail Richness in Text-to-3D (66 citations)
- Symphonize 3D Semantic Scene Completion with Contextual Instance Queries (60 citations)
and 157 more in the interactive map.
Face and shape · 126 papers
- GaussianAvatars: Photorealistic Head Avatars with Rigged 3D Gaussians (241 citations)
- GaussianAvatar: Towards Realistic Human Avatar Modeling from a Single Video via Animatable 3D Gaussians (171 citations)
- MeshGPT: Generating Triangle Meshes with Decoder-Only Transformers (113 citations)
- Gaussian Head Avatar: Ultra High-fidelity Head Avatar via Dynamic Gaussians (109 citations)
- Relightable Gaussian Codec Avatars (102 citations)
- DreamAvatar: Text-and-Shape Guided 3D Human Avatar Generation via Diffusion Models (100 citations)
- GART: Gaussian Articulated Template Models (99 citations)
- FlashAvatar: High-fidelity Head Avatar with Efficient Gaussian Embedding (77 citations)
- SyncTalk: The Devil is in the Synchronization for Talking Head Synthesis (76 citations)
- Paint3D: Paint Anything 3D with Lighting-Less Texture Diffusion Models (64 citations)
- SIFU: Side-view Conditioned Implicit Function for Real-world Usable Clothed Human Reconstruction (57 citations)
- SiTH: Single-view Textured Human Reconstruction with Image-Conditioned Diffusion (52 citations)
and 114 more in the interactive map.
Depth and stereo · 118 papers
- DUSt3R: Geometric 3D Vision Made Easy (545 citations)
- SplaTAM: Splat Track & Map 3D Gaussians for Dense RGB-D SLAM (438 citations)
- Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation (388 citations)
- Photo-SLAM: Real-time Simultaneous Localization and Photorealistic Mapping for Monocular Stereo and RGB-D Cameras (205 citations)
- UniDepth: Universal Monocular Metric Depth Estimation (189 citations)
- Depth Information Assisted Collaborative Mutual Promotion Network for Single Image Dehazing (165 citations)
- MoCha-Stereo: Motif Channel Attention Network for Stereo Matching (86 citations)
- SNI-SLAM: Semantic Neural Implicit SLAM (82 citations)
- VGGSfM: Visual Geometry Grounded Deep Structure From Motion (81 citations)
- PLGSLAM: Progressive Neural Scene Represenation with Local to Global Bundle Adjustment (70 citations)
- SelfOcc: Self-Supervised Vision-Based 3D Occupancy Prediction (65 citations)
- Detector-Free Structure from Motion (59 citations)
and 106 more in the interactive map.
Human pose estimation · 93 papers
- FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects (315 citations)
- Reconstructing Hands in 3D with Transformers (151 citations)
- KTPFormer: Kinematics and Trajectory Prior Knowledge-Enhanced Transformer for 3D Human Pose Estimation (107 citations)
- WHAM: Reconstructing World-grounded Humans with Accurate 3D Motion (98 citations)
- RTMO: Towards High-Performance One-Stage Real-Time Multi-Person Pose Estimation (97 citations)
- GigaPose: Fast and Robust Novel Object Pose Estimation via One Correspondence (87 citations)
- TokenHMR: Advancing Human Mesh Recovery with a Tokenized Pose Representation (70 citations)
- Hourglass Tokenizer for Efficient Transformer-Based 3D Human Pose Estimation (67 citations)
- FinePOSE: Fine-Grained Prompt-Driven 3D Human Pose Estimation via Diffusion Models (55 citations)
- ChatPose: Chatting about 3D Human Pose (54 citations)
- A Dual-Augmentor Framework for Domain Generalization in 3D Human Pose Estimation (49 citations)
- 6D-Diff: A Keypoint Diffusion Framework for 6D Object Pose Estimation (39 citations)
and 81 more in the interactive map.
NeRF and radiance fields · 73 papers
- ReconFusion: 3D Reconstruction with Diffusion Priors (185 citations)
- DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Vision (140 citations)
- Structure-Aware Sparse-View X-ray 3D Reconstruction (77 citations)
- NeuRAD: Neural Rendering for Autonomous Driving (70 citations)
- Bayes' Rays: Uncertainty Quantification for Neural Radiance Fields (58 citations)
- GARField: Group Anything with Radiance Fields (53 citations)
- NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild (48 citations)
- How Far Can We Compress Instant-NGP-Based NeRF? (34 citations)
- MuRF: Multi-Baseline Radiance Fields (34 citations)
- Neural Underwater Scene Representation (32 citations)
- PIE-NeRF: Physics-based Interactive Elastodynamics with NeRF (31 citations)
- Tactile-Augmented Radiance Fields (28 citations)
and 61 more in the interactive map.
Gaussian splatting · 52 papers
- 4D Gaussian Splatting for Real-Time Dynamic Scene Rendering (737 citations)
- Mip-Splatting: Alias-free 3D Gaussian Splatting (524 citations)
- SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering (478 citations)
- Scaffold-GS: Structured 3D Gaussians for View-Adaptive Rendering (463 citations)
- Gaussian Splatting SLAM (461 citations)
- Deformable 3D Gaussians for High-Fidelity Monocular Dynamic Scene Reconstruction (420 citations)
- GS-SLAM: Dense Visual SLAM with 3D Gaussian Splatting (360 citations)
- pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction (273 citations)
- Compact 3D Gaussian Representation for Radiance Field (257 citations)
- LangSplat: 3D Language Gaussian Splatting (248 citations)
- DrivingGaussian: Composite Gaussian Splatting for Surrounding Dynamic Autonomous Driving Scenes (223 citations)
- Feature 3DGS: Supercharging 3D Gaussian Splatting to Enable Distilled Feature Fields (215 citations)
and 40 more in the interactive map.