TPAMI 2026 papers on 3D vision
82 of the 880 papers published in TPAMI 2026 (9.3%) are on 3D vision, compared with 15.5% at TPAMI 2025. They fall into 6 topics.
See TPAMI 2026 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
Depth and stereo · 19 papers
- RENI++: A Rotation-Equivariant, Scale-Invariant, Natural Illumination Prior (11 citations)
- Revisiting 360 Depth Estimation With PanoGabor: A New Fusion Perspective (6 citations)
- Single-Photon Imaging in Complex Scenarios via Physics-Informed Deep Neural Networks (5 citations)
- Discriminately Treating Motion Components Evolves Joint Depth and Ego-Motion Learning (4 citations)
- A Hierarchical Prior Mining Approach for Non-Local Multi-View Stereo (2 citations)
- Pixel-Perfect Visual Geometry Estimation (2 citations)
- A Minimal Solution to the Perspective-3-Point Problem for the Camera With Unknown Focal Length and Two Degrees-of-Freedom Rotation (1 citation)
- Robust Model Fitting via Motion-Aware Pyramid Transformer-Guided Preference Filtering and Consensus Smoothing (1 citation)
- Aperture-aware Dispersion 5-D Light-field Imaging Spectrometer
- Digging Into Monocular Texture-Scale Feature Fusion for Optical Flow and Optical Expansion Estimation
- DP-SfM: Dual-Pixel Structure-From-Motion Without Scale Ambiguity
- EAR-Net: Pursuing End-to-End Absolute Rotations From Multi-View Images
and 7 more in the interactive map.
3D generation · 18 papers
- OmniHD-Scenes: A Next-Generation Multimodal Dataset for Autonomous Driving (30 citations)
- Scaling up Occupancy-Centric Driving Scene Generation: Dataset and Method (9 citations)
- Hierarchical Context Alignment With Disentangled Geometric and Temporal Modeling for Semantic Occupancy Prediction (7 citations)
- Robust 3D Semantic Occupancy Prediction With Calibration-Free Spatial Transformation (7 citations)
- Hermes++: Toward a Unified Driving World Model for 3D Scene Understanding and Generation (5 citations)
- ArtHOI: Articulated Human-Object Interaction Synthesis by 4D Reconstruction From Video Priors (2 citations)
- Chat-Scene++: Exploiting Context-Rich Object Identification for 3D LLM (2 citations)
- OneBEV++: Towards Unifying Bird's-Eye-View Semantic Mapping with Panoramas (2 citations)
- Learning Diffusion Priors for Inverse Rendering Under Unknown Illumination (1 citation)
- MaterialMVP++: High-Quality Material Generation With Neural Bump Estimation Via Multi-View PBR Diffusion (1 citation)
- Unifying Multi-Modal Hair Editing via Proxy Feature Blending (1 citation)
- 4DWeaver: Bridging Reconstruction and Generation Via Compact Autoregressive Priors
and 6 more in the interactive map.
Gaussian splatting · 17 papers
- Learning-Based Multi-View Stereo: A Survey (46 citations)
- A Survey on 3D Gaussian Splatting Applications: Segmentation, Editing, and Generation (32 citations)
- LangSurf: Language-Embedded Surface Gaussians for 3D Scene Understanding (27 citations)
- FreeSplat++: Generalizable 3D Gaussian Splatting for Efficient Indoor Scene Reconstruction (16 citations)
- DrivingGaussian++: Toward Realistic Reconstruction and Editable Simulation for Surrounding Dynamic Driving Scenes (7 citations)
- DNGaussian++: Improving Sparse-View Gaussian Radiance Fields With Depth Normalization (3 citations)
- On the Design of Mixture-of-Experts for Dynamic Gaussian Splatting (3 citations)
- Thermal3D-GS: Physics-Induced 3D Gaussians for Thermal Infrared Novel-View Synthesis With a Large-Scale Dataset (2 citations)
- DroneSplat+: Semantics-Enhanced 3D Gaussian Splatting for Robust 3D Reconstruction From In-the-Wild Drone Imagery (1 citation)
- RealLiFe: Real-Time Light Field Reconstruction via Hierarchical Sparse Gradient Descent (1 citation)
- ThermalGaussian++: Improving Alignment and Resolution for ThermalGaussian (1 citation)
- TraGraph-GS: Trajectory Graph-based Gaussian Splatting for Arbitrary Large-Scale Scene Rendering (1 citation)
and 5 more in the interactive map.
Human pose estimation · 13 papers
- MVHumanNet++: A Large-Scale Dataset of Multi-View Daily Dressing Human Captures With Richer Annotations for 3D Human Digitization (11 citations)
- Human Motion Prediction via Continual Prior Compensation (8 citations)
- A Comparative Assessment of Accuracy in Video-Based Monocular Human Pose Estimation Frameworks (6 citations)
- A Pose-Only Geometric Constraint for Multi-Camera Pose Adjustment (5 citations)
- Probing Effective and Efficient Category-Level Articulated Object Pose Perception (2 citations)
- Uni-Hand: Universal Hand Motion Forecasting in Egocentric Views (2 citations)
- Scalable and Generalizable Correspondence Pruning via Geometry-Consistent Pre-Training (1 citation)
- GVHMR: Gravity-View Coordinates for Global Human Motion Recovery From Monocular Videos
- OptimalCap: Efficient and Robust LiDAR-Based Motion Capture in Free Environments
- Person-in-WiFi 3D: Unified Model for 3D WiFi Perception
- Transformers With Joint Tokens and Local-Global Attention for Efficient Human Pose Estimation
- Unleashing the Power of Text-to-Image Diffusion Models for Category-Agnostic Pose Estimation
and 1 more in the interactive map.
Face and shape · 12 papers
- Cross-Frequency Implicit Neural Representation With Self-Evolving Parameters (3 citations)
- Interacted Planes Reveal 3D Line Mapping (2 citations)
- Motion2VecSets: Non-Rigid Shape Reconstruction and Tracking With 4D Latent Set Diffusion (1 citation)
- StdGEN++: A Comprehensive System for Semantic-Decomposed 3D Character Generation (1 citation)
- UDFStudio: A Unified Framework of Datasets, Benchmarks and Generative Models for Unsigned Distance Functions (1 citation)
- A Riemannian Framework for the Elastic Analysis of the Spatiotemporal Variability in the Shape and Structure of Tree-Like 4D Objects
- Hierarchical Mesh Representation Learning With Spectral Dictionary Embedding
- QSMnet-INR: Single-Orientation Quantitative Susceptibility Mapping via Implicit Neural Representation in k-Space
- Spatio-Temporal Garment Reconstruction Using Diffusion Mapping via Pattern Coordinates
- Study of Full-View Finger Vein Biometrics on Redundancy-Compensated 3D Reconstruction and Frequency-Spatial Coupling Feature Extraction
- FINER++: Building a Family of Variable-periodic Functions for Activating Implicit Neural Representation
- VRP-UDF: Toward Unbiased Learning of Unsigned Distance Functions From Multi-View Images With Volume Rendering Priors
