Niagara : Normal-Integrated Geometric Affine Field for Scene Reconstruction From a Single View
- 12 December 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (5) , 6866-6876
Abstract
Recent advances in single-view 3D scene reconstruction have highlighted the challenges in capturing fine geometric details and ensuring structural consistency, particularly in high-fidelity outdoor scene modeling. This paper presents Niagara, a new single-view 3D scene reconstruction framework that can faithfully reconstruct challenging outdoor scenes from a single input image for the first time. Our approach integrates monocular depth and normal estimation as input, which substantially improves its ability to capture fine details, mitigating common issues like geometric detail loss and deformation. Additionally, we introduce a geometric affine field (GAF) and 3D self-attention as geometry-constraint, which combines the structural properties of explicit geometry with the adaptability of implicit feature fields, striking a balance between efficient rendering and high-fidelity reconstruction. Our framework finally proposes a specialized encoder-decoder architecture, where a depth-based 3D Gaussian decoder is proposed to predict 3D Gaussian parameters, which can be used for novel view synthesis. Extensive results and analyses suggest that our Niagara surpasses prior SoTA approaches such as Flash3D in both single-view and dual-view settings, significantly enhancing the geometric accuracy and visual fidelity, especially in outdoor scenes. Webpage: https://ai-kunkun.github.io/Niagara page/.Keywords
All Related Versions
Funding Information
- Youth Science Fund Project (62506305)
- Zhejiang Leading Innovative and Entrepreneur Team Introduction Program (2024R01007)
- Key Research and Development Program of Zhejiang Province (2025C01026)
- Scientific Research Project of Westlake University (WU2025WF003)
- National University Research Fund of China (2025CAAI-ANT-13)
This publication has 60 references indexed in Scilit:
- Flash3D: Feed-Forward Generalisable 3D Scene Reconstruction from a Single Image2025
- LatentSplat: Autoencoding Variational Gaussians for Fast Generalizable 3D Reconstruction2024
- MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-view Images2024
- TensoRF: Tensorial Radiance Fields2022
- R2L: Distilling Neural Radiance Field to Neural Light Field for Efficient Novel View Synthesis2022
- MVSNet: Depth Inference for Unstructured Multi-view Stereo2018
- Indoor Segmentation and Support Inference from RGBD Images2012
- A Comparison and Evaluation of Multi-View Stereo Reconstruction Algorithms2006
- A cooperative algorithm for stereo matching and occlusion detectionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2000
- Optical models for direct volume renderingIEEE Transactions on Visualization and Computer Graphics, 1995