End-to-End-3D-Reconstruction-Paper-List

Personal list. With relevant research advancing fast and branching out widely, I'll only add papers meeting my needs hereafter.

---

• 3D Reconstruction
• Scalable
• Self Supervised
• Semantic
• Dynamic
• Generation
• Novel View Synthesis

---

3D Reconstruction

• DUSt3R: Geometric 3D Vision Made Easy [CVPR 2024] [dust3r]

• Grounding Image Matching in 3D with MASt3R [ECCV 2024] [mast3r]

• 3D Reconstruction with Spatial Memory [3DV 2025] [spann3r]

• Continuous 3D Perception Model with Persistent State [CVPR 2025] [cut3r]

• Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Pass [CVPR 2025] [fast3r-3d]

• MUSt3R: Multi-view Network for Stereo 3D Reconstruction [CVPR 2025] [must3r]

• PE3R: Perception-Efficient 3D Reconstruction [CVPR 2026] [pe3r]

• VGGT: Visual Geometry Grounded Transformer [CVPR 2025] [vggt]

• Pow3R: Empowering Unconstrained 3D Reconstruction with Camera and Scene Priors [CVPR 2025] []

• Matrix3D: Large Photogrammetry Model All-in-One [CVPR 2025] [ml-matrix3d]

• DiffusionSfM: Predicting Structure and Motion via Ray Origin and Endpoint Diffusion [CVPR 2025] [DiffusionSfM]

• Point3R: Streaming 3D Reconstruction with Explicit Spatial Pointer Memory [NeurIPS 2025] [Point3R]

• π³: Scalable Permutation-Equivariant Visual Geometry Learning [ICLR 2026] [Pi3]

• StreamVGGT: Streaming 4D Visual Geometry Transformer [ICLR 2026] [StreamVGGT]

• STream3R: Scalable Sequential 3D Reconstruction with Causal Transformer [ICLR 2026] [STream3R]

• WinT3R: Window-Based Streaming Reconstruction With Camera Token Pool [arXiv 2025] [WinT3R]

• SAIL-Recon: Large SfM by Augmenting Scene Regression with Localization [3DV 2026] [sail-recon]

• FastVGGT: Training-Free Acceleration of Visual Geometry Transformer [ICLR 2026] [FastVGGT]

• Faster VGGT with Block-Sparse Global Attention [CVPR 2026] [sparse-vggt]

• Quantized Visual Geometry Grounded Transformer [ICLR 2026] [QuantVGGT]

• MapAnything: Universal Feed-Forward Metric 3D Reconstruction [3DV 2026] [map-anything]

• TTT3R: 3D Reconstruction as Test-Time Training [ICLR 2026] [TTT3R]

• WorldMirror: Universal 3D World Reconstruction with Any-Prior Prompting [arXiv 2025] [HunyuanWorld-Mirror]

• OmniVGGT: Omni-Modality Driven Visual Geometry Grounded Transformer [CVPR 2026] [OmniVGGT]

• Depth Anything 3: Recovering the Visual Space from Any Views [ICLR 2026] [DA3]

• NOVA3R: Non-pixel-aligned Visual Transformer for Amodal 3D Reconstruction [ICLR 2026] [nova3r]

• AMB3R: Accurate Feed-forward Metric-scale 3D Reconstruction with Backend [CVPR 2026] [amb3r]

• HD-VGGT: High-Resolution Visual Geometry Transformer [arXiv 2026] []

• DAGE: Dual-Stream Architecture for Efficient and Fine-Grained Geometry Estimation [CVPR 2026] [DAGE]

• Unlocking the Power of Critical Factors for 3D Visual Geometry Estimation [CVPR 2026] [CARVE]

• VGGT-Ω [CVPR 2026] [vggt-omega]

• GenRecon: Bridging Generative Priors for Multi-View 3D Scene Reconstruction [arXiv 2026] [GenRecon]

• Déjà View: Looping Transformers for Multi-View 3D Reconstruction [arXiv 2026] [dvlt]

• Surflo: Consistent 3D Surface Flow Model with Global State [arXiv 2026] [surflo]

• 🌐 Argus: Metric Panoramic 3D Reconstruction for Indoor Scenes [ECCV 2026] [argus]

Scalable

• Offline Feed-Forward 3D Reconstruction at Scale [arXiv 2026] []

• ZipMap: Linear-Time Stateful 3D Reconstruction via Test-Time Training [CVPR 2026] [ZipMap]

• LoGeR: Long-Context Geometric Reconstruction with Hybrid Memory [arXiv 2026] [LoGeR]

• S-VGGT: Structure-Aware Subscene Decomposition for Scalable 3D Foundation Models [ICME 2026] [S-VGGT]

• Scal3R: Scalable Test-Time Training for Large-Scale 3D Reconstruction [CVPR 2026] [scal3r]

• LingBot-Map: Geometric Context Transformer for Streaming 3D Reconstruction [arXiv 2026] [lingbot-map]

• MERG3R: A Divide-and-Conquer Approach to Large-Scale Neural Visual Geometry [CVPR 2026] [MERG3R]

• HorizonStream: Long-Horizon Attention for Streaming 3D Reconstruction [arXiv 2026] [HorizonStream]

• Diversity-aware View Partitioning for Scalable VGGT [ECCV 2026] [DA-VGGT]

Self-Supervised

• Reliev3R: Relieving Feed-forward 3D Reconstruction from Multi-View Geometric Annotations [arXiv 2026] []

• From None to All: Self-Supervised 3D Reconstruction via Novel View Synthesis [CVPR 2026] [nas3r]

• FF3R: Feedforward Feature 3D Reconstruction from Unconstrained views [CVPRF 2026] [ff3r]

Semantic

• PanSt3R: Multi-view Consistent Panoptic Segmentation [ICCV 2025] [panst3r]

• IGGT: Instance-Grounded Geometry Transformer for Semantic 3D Reconstruction [ICLR 2026] [IGGT_official]

• Uni3R: Unified 3D Reconstruction and Semantic Understanding via Generalizable Gaussian Splatting from Unposed Multi-View Images [CVPR 2026] [Uni3R]

• VGGT-Segmentor: Geometry-Enhanced Cross-View Segmentation [arXiv 2026] []

• EPS3D : End-to-End Feed-Forward 3D Panoptic Segmentation [ICML 2026] [EPS3D]

• SegVGGT: Joint 3D Reconstruction and InstanceSegmentation from Multi-View Images [ECCV 2026] [SegVGGT]

Dynamic

• MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motion [ICLR 2025] [monst3r]

• MegaSaM: Accurate, Fast, and Robust Structure and Motion from Casual Dynamic Videos [CVPR 2025] [mega-sam]

• Driv3R: Learning Dense 4D Reconstruction for Autonomous Driving [arXiv 2024] [Driv3R]

• Geo4D: Leveraging Video Generators for Geometric 4D Scene Reconstruction [ICCV 2025] [Geo4D]

• ViPE: Video Pose Engine for Geometric 3D Perception [arXiv 2025] [vipe]

• VGGT4D: Mining Motion Cues in Visual Geometry Transformers for 4D Scene Reconstruction [arXiv 2025] [vggt4d]

• Efficiently Reconstructing Dynamic Scenes One D4RT at a Time [CVPR 2026] [d4rt]

Generation

• ReconViaGen: Towards Accurate Multi-view 3D Object Reconstruction via Generation [ICLR 2026] [ReconViaGen]

• UniRecGen: Unifying Multi-View 3D Reconstruction and Generation [arXiv 2025] [UniRecGen]

• Stepper: Stepwise Immersive Scene Generation with Multiview Panorama [CVPRF 2026] [stepper]

• VGGRPO: Towards World-Consistent Video Generation with 4D Latent Reward [arXiv 2026] [VGGRPO]

• Latent Riemannian Flow Matching for Geometry-Grounded 3D Foundation Models [arXiv 2026] [geometry-grounded-rfm]

• PixWorld: Unifying 3D Scene Generation and Reconstruction in Pixel Space [arXiv 2026] [PixWorld]

• GenRec: Knowing Where to Reconstruct and Where to Generate [arXiv 2026] [GenRec]

Novel View Synthesis

• Splatt3R: Zero-shot Gaussian Splatting from Uncalibrated Image Pairs [arXiv 2024] [splatt3r]

• No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images [ICLR 2025] [NoPoSplat]

• PreF3R: Pose-Free Feed-Forward 3D Gaussian Splatting from Variable-length Image Sequence [arXiv 2024] [PreF3R]

• SPARS3R: Semantic Prior Alignment and Regularization for Sparse 3D Reconstruction [CVPR 2025] [SPARS3R]

• LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias [ICLR 2025] [LVSM]

• FlowR: Flowing from Sparse to Dense 3D Reconstructions [ICCV 2025] [flowr]

• RayZer: A Self-supervised Large View Synthesis Model [ICCV 2025] [RayZer]

• AnySplat: Feed-forward 3D Gaussian Splatting from Unconstrained Views [SIGGRAPH Asia 2025] [AnySplat]

• VGGT-X: When VGGT Meets Dense Novel View Synthesis [arXiv 2025] [VGGT-X]

• YoNoSplat: You Only Need One Model for Feedforward 3D Gaussian Splatting [ICLR 2026] [yonosplat]

• E-RayZer: Self-supervised 3D Reconstruction as Spatial Visual Pre-training [CVPR 2026] [E-RayZer]

• Off The Grid: Detection of Primitives for Feed-Forward 3D Gaussian Splatting [CVPR 2026] [OffTheGrid]

• Sharp Monocular View Synthesis in Less Than a Second [arXiv 2025] [ml-sharp]

• UniSHARP: Universal Sharp Monocular View Synthesis [arXiv 2026] [UniSHARP]

• EcoSplat: Efficiency-controllable Feed-forward 3D Gaussian Splatting from Multi-view Images [CVPR 2026] [ecosplat-site]

• From Rays to Projections: Better Inputs for Feed-Forward View Synthesis [CVPR 2026] [pvsm-web]

• Diff3R: Feed-forward 3D Gaussian Splatting with Uncertainty-aware Differentiable Optimization [arXiv 2026] [diff3r]

• One-Shot Refiner: Boosting Feed-forward Novel View Synthesis via One-Step Diffusion [arXiv 2026] [One_Shot_Refiner]

• Repurposing Geometric Foundation Models for Multi-view Diffusion [arXiv 2026] [GLD]

• UniQueR: Unified Query-based Feedforward 3D Reconstruction [arXiv 2026] []

• Leveling3D: Leveling Up 3D Reconstruction withFeed-Forward 3D Gaussian Splatting andGeometry-Aware Generation [arXiv 2026] []

• Less Gaussians, Texture More: 4K Feed-Forward Textured Splatting [ICLR 2026] [lgtm]

• LagerNVS: Latent Geometry for Fully Neural Real-Time Novel View Synthesis [CVPR 2026] [lagernvs]

• Pose-Free Omnidirectional Gaussian Splatting for 360-Degree Videos with Consistent Depth Priors [CVPR 2026] [PFGS360]

• FreeScale: Scaling 3D scenes via Certainty-Aware Free-View Generation [CVPR 2026] [FreeScale]

• C3G: Learning Compact 3D Representations with 2K Gaussians [CVPR 2026] [C3G]

• ZipSplat: Fewer Gaussians, Better Splats [arXiv 2026] [ZipSplat]

• AnyRecon: Arbitrary-View 3D Reconstruction with Video Diffusion Model [arXiv 2026] [AnyRecon]

• QuerySplat: Decoupling Geometry and Appearance Representations in 3DGS Prediction [arXiv 2026] [querysplat]