AAAI 2025 accepted papers on 3D vision
140 of the 3,486 papers accepted at AAAI 2025 (4.0%) are on 3D vision, compared with 4.0% at AAAI 2024. They fall into 6 topics.
See AAAI 2025 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 · 39 papers
- DME-Driver: Integrating Human Decision Logic and 3D Scene Perception in Autonomous Driving (21 citations)
- CAD-GPT: Synthesising CAD Construction Sequence with Spatial Reasoning-Enhanced Multimodal LLMs (20 citations)
- MagicMan: Generative Novel View Synthesis of Humans with 3D-Aware Diffusion and Iterative Refinement (20 citations)
- MV-VTON: Multi-View Virtual Try-On with Diffusion Models (18 citations)
- Cycle3D: High-quality and Consistent Image-to-3D Generation via Generation-Reconstruction Cycle (16 citations)
- SceneX: Procedural Controllable Large-Scale Scene Generation (13 citations)
- MMGDreamer: Mixed-Modality Graph for Geometry-Controllable 3D Indoor Scene Generation (11 citations)
- OmniSR: Shadow Removal Under Direct and Indirect Lighting (9 citations)
- VLScene: Vision-Language Guidance Distillation for Camera-Based 3D Semantic Scene Completion (8 citations)
- DriveEditor: A Unified 3D Information-Guided Framework for Controllable Object Editing in Driving Scenes (6 citations)
- Mamba-CAD: State Space Model for 3D Computer-Aided Design Generative Modeling (6 citations)
- Sequence Matters: Harnessing Video Models in 3D Super-Resolution (5 citations)
and 27 more in the interactive map.
Gaussian splatting · 29 papers
- Segment Any 3D Gaussians (75 citations)
- TranSplat: Generalizable 3D Gaussian Splatting from Sparse Multi-View Images with Transformers (28 citations)
- DrivingForward: Feed-forward 3D Gaussian Splatting for Driving Scene Reconstruction from Flexible Surround-view Input (14 citations)
- FatesGS: Fast and Accurate Sparse-View Surface Reconstruction Using Gaussian Splatting with Depth-Feature Consistency (14 citations)
- GGS: Generalizable Gaussian Splatting for Lane Switching in Autonomous Driving (9 citations)
- Large Images Are Gaussians: High-Quality Large Image Representation with Levels of 2D Gaussian Splatting (9 citations)
- Efficient Gaussian Splatting for Monocular Dynamic Scene Rendering via Sparse Time-Variant Attribute Modeling (8 citations)
- GigaGS: 3D Gaussian Based Planar Representation for Large-Scene Surface Reconstruction (8 citations)
- Frequency-Aware Density Control via Reparameterization for High-Quality Rendering of 3D Gaussian Splatting (7 citations)
- GraphAvatar: Compact Head Avatars with GNN-Generated 3D Gaussians (7 citations)
- DeRainGS: Gaussian Splatting for Enhanced Scene Reconstruction in Rainy Environments (6 citations)
- EvHDR-GS: Event-guided HDR Video Reconstruction with 3D Gaussian Splatting (6 citations)
and 17 more in the interactive map.
Depth and stereo · 22 papers
- GFlow: Recovering 4D World from Monocular Video (18 citations)
- DVP-MVS: Synergize Depth-Edge and Visibility Prior for Multi-View Stereo (16 citations)
- MSP-MVS: Multi-Granularity Segmentation Prior Guided Multi-View Stereo (13 citations)
- RRT-MVS: Recurrent Regularization Transformer for Multi-View Stereo (7 citations)
- DeMo: Deep Motion Field Consensus with Learnable Kernels for Two-view Correspondence Learning (5 citations)
- DiffCalib: Reformulating Monocular Camera Calibration as Diffusion-Based Dense Incident Map Generation (5 citations)
- Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization (4 citations)
- FMPM-DNet: Hyperspectral Pansharpening Dynamic Network Based on Feature Modulation and Probability Mask (4 citations)
- Intrinsic Image Decomposition for Robust Self-supervised Monocular Depth Estimation on Reflective Surfaces (3 citations)
- Sparis: Neural Implicit Surface Reconstruction of Indoor Scenes from Sparse Views (3 citations)
- Dual-branch Graph Feature Learning for NLOS Imaging (2 citations)
- SVRMamba: Slice-to-Volume Reconstruction from Multiple MRI Stacks with Slice Sequence Guided Mamba (2 citations)
and 10 more in the interactive map.
Human pose estimation · 20 papers
- TCPFormer: Learning Temporal Correlation with Implicit Pose Proxy for 3D Human Pose Estimation (35 citations)
- PoseMamba: Monocular 3D Human Pose Estimation with Bidirectional Global-Local Spatio-Temporal State Space Model (29 citations)
- GenHMR: Generative Human Mesh Recovery (8 citations)
- Pose Magic: Efficient and Temporally Consistent Human Pose Estimation with a Hybrid Mamba-GCN Network (8 citations)
- MUC: Mixture of Uncalibrated Cameras for Robust 3D Human Body Reconstruction (6 citations)
- Multi-View 3D Human Pose Estimation with Weakly Synchronized Images (4 citations)
- SpatioTemporal Learning for Human Pose Estimation in Sparsely-Labeled Videos (4 citations)
- SVTformer: Spatial-View-Temporal Transformer for Multi-View 3D Human Pose Estimation (4 citations)
- Optimizing Human Pose Estimation Through Focused Human and Joint Regions (3 citations)
- Causal-Inspired Multitask Learning for Video-Based Human Pose Estimation (2 citations)
- Free-Moving Object Reconstruction and Pose Estimation with Virtual Camera (2 citations)
- FreeCap: Hybrid Calibration-Free Motion Capture in Open Environments (2 citations)
and 8 more in the interactive map.
Face and shape · 18 papers
- Details Enhancement in Unsigned Distance Field Learning for High-fidelity 3D Surface Reconstruction (12 citations)
- DEEPTalk: Dynamic Emotion Embedding for Probabilistic Speech-Driven 3D Face Animation (11 citations)
- Adaptive Wavelet-Positional Encoding for High-Frequency Information Learning in Implicit Neural Representation (6 citations)
- Do Not DeepFake Me: Privacy-Preserving Neural 3D Head Reconstruction Without Sensitive Images (4 citations)
- Enhancing Implicit Neural Representations via Symmetric Power Transformation (3 citations)
- Textured Mesh Saliency: Bridging Geometry and Texture for Human Perception in 3D Graphics (3 citations)
- 3D²-Actor: Learning Pose-Conditioned 3D-Aware Denoiser for Realistic Gaussian Avatar Modeling (2 citations)
- High-Fidelity Polarimetric Implicit 3D Reconstruction with View-Dependent Physical Representation (2 citations)
- Mesh Watermark Removal Attack and Mitigation: A Novel Perspective of Function Space (2 citations)
- CtrlAvatar: Controllable Avatars Generation via Disentangled Invertible Networks (1 citation)
- Hierarchically Controlled Deformable 3D Gaussians for Talking Head Synthesis (1 citation)
- IPVTON: Image-based 3D Virtual Try-on with Image Prompt Adapter (1 citation)
and 6 more in the interactive map.
NeRF and radiance fields · 12 papers
- AE-NeRF: Augmenting Event-Based Neural Radiance Fields for Non-ideal Conditions and Larger Scenes (11 citations)
- EvHDR-NeRF: Building High Dynamic Range Radiance Fields with Single Exposure Images and Events (7 citations)
- CodecNeRF: Toward Fast Encoding and Decoding, Compact, and High-quality Novel-view Synthesis (4 citations)
- Rethinking Open-Vocabulary Segmentation of Radiance Fields in 3D Space (3 citations)
- Bright-NeRF: Brightening Neural Radiance Field with Color Restoration from Low-Light RAW Images (2 citations)
- Spatial Annealing for Efficient Few-shot Neural Rendering (2 citations)
- Spatiotemporal-Aware Neural Fields for Dynamic CT Reconstruction (2 citations)
- Decoupling Scattering: Pseudo-Label Guided NeRF for Scenes with Scattering Media (1 citation)
- Physical-aware Neural Radiance Fields for Efficient Exposure Correction (1 citation)
- STGC-NeRF: Spatial-Temporal Geometric Consistency for LiDAR Neural Radiance Fields in Dynamic Scenes (1 citation)
- DetRF: Detachable Novel Views Synthesis of Dynamic Scenes Using Backdrop-Driven Neural Radiance Fields
- Normal-NeRF: Ambiguity-Robust Normal Estimation for Highly Reflective Scenes