由于近场与远场区域的监督信号不一致，存在极端深度变化场景的三维重建仍具挑战性。现有方法无法同时解决远距离区域深度估计不准确和近距离区域结构退化的问题。本文提出一种融合景深监督与多视图一致性监督的新型计算框架，以改进3D Gaussian Splatting。该方法包含两个核心组件：（1）景深监督模块采用尺度恢复的单目深度估计器（如Metric3D）生成深度先验，利用散焦卷积合成物理准确的散焦图像，并通过新型景深损失强制几何一致性，从而提升远场与近场区域的深度保真度；（2）多视图一致性监督模块采用基于LoFTR的半稠密特征匹配最小化跨视图几何误差，并通过可靠匹配点的最小二乘优化强制深度一致性。通过统一散焦物理与多视图几何约束，该方法实现了更优的深度保真度，在Waymo Open Dataset上相比当前最优方法取得了0.8 dB的PSNR提升。该框架架起了物理成像原理与基于学习的深度正则化之间的桥梁，为城市环境中的复杂深度分层问题提供了可扩展解决方案。
英文摘要：
Three-dimensional reconstruction in scenes with extreme depth variations remains challenging due to inconsistent supervisory signals between near-field and far-field regions. Existing methods fail to simultaneously address inaccurate depth estimation in distant areas and structural degradation in close-range regions. This paper proposes a novel computational framework that integrates depth-of-field supervision and multi-view consistency supervision to advance 3D Gaussian Splatting. Our approach comprises two core components: (1) Depth-of-field Supervision employs a scale-recovered monocular depth estimator (e.g., Metric3D) to generate depth priors, leverages defocus convolution to synthesize physically accurate defocused images, and enforces geometric consistency through a novel depth-of-field loss, thereby enhancing depth fidelity in both far-field and near-field regions; (2) Multi-View Consistency Supervision employing LoFTR-based semi-dense feature matching to minimize cross-view geometric errors and enforce depth consistency via least squares optimization of reliable matched points. By unifying defocus physics with multi-view geometric constraints, our method achieves superior depth fidelity, demonstrating a 0.8 dB PSNR improvement over the state-of-the-art method on the Waymo Open Dataset. This framework bridges physical imaging principles and learning-based depth regularization, offering a scalable solution for complex depth stratification in urban environments.
DOF-GS:Adjustable Depth-of-Field 3D Gaussian Splatting for Post-Capture Refocusing, Defocus Rendering and Blur Removal
State Key Laboratory of General Artificial Intelligence, Peking University（通用人工智能国家重点实验室，北京大学）；School of Intelligence Science and Technology, Peking University（智能科学与技术学院，北京大学）；University of North Carolina at Chapel Hill（北卡罗来纳大学教堂山分校）
ClearGS: Reliability-Aware Gaussian Splatting from Handheld Videos
Shenzhen University of Advanced Technology（深圳理工大学）；HONOR（荣耀）；University of New South Wales（新南威尔士大学）；Southern University of Science and Technology（南方科技大学）
GSPotential: Camera Potential Field for Sparse-View 3D Gaussian Splatting
State Key Laboratory of Virtual Reality Technology and Systems, Beihang University（北京航空航天大学虚拟现实技术与系统国家重点实验室）；School of Computer Science and Engineering, Beihang University（北京航空航天大学计算机科学与工程学院）；Zhongguancun Laboratory（中关村实验室）
Data-driven depth estimation methods struggle with the generalization outside their training scenes due to the immense variability of the real-world scenes. This problem can be partially addressed by utilising synthetically generated images, but closing the synthetic-real domain gap is far from trivial. In this paper, we tackle this issue by using domain invariant defocus blur as direct supervision. We leverage defocus cues by using a permutation invariant convolutional neural network that encourages the network to learn from the differences between images with a different point of focus. Our …
ERGO: Excess-Risk-Guided Optimization for High-Fidelity Monocular 3D Gaussian Splatting
Shenzhen Campus of Sun Yat-sen University（中山大学深圳校区）；Peking University（北京大学）；Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）；Guilin University of Electronic Technology（桂林电子科技大学）；UiT The Arctic University of Norway（UiT北欧大学）
GA-GS: Generation-Assisted Gaussian Splatting for Static Scene Reconstruction
School of Computer Science and Technology, University of Science and Technology of China（中国科学技术大学计算机科学与技术学院）；School of Artificial Intelligence and Data Science, University of Science and Technology of China（中国科学技术大学人工智能与数据科学学院）；The Chinese University of Hong Kong（香港中文大学）；Institute of Advanced Technology, University of Science and Technology of China（中国科学技术大学先进技术研究院）；Institute of Artificial Intelligence, Hefei Comprehensive National Science Center（合肥综合性国家科学中心人工智能研究院）；School of information Science and Technology, University of Science and Technology of China（中国科学技术大学信息科学技术学院）
HarmoGS: Robust 3D Gaussian Splatting in the Wild via Conflict-Aware Gradient Harmonization
Sun Yat-sen University（中山大学）；Northeastern University（东北大学）；Guangdong Province Key Laboratory of Information Security Technology（广东省信息安全技术重点实验室）；Key Laboratory of Machine Intelligence and Advanced Computing, Ministry of Education, China（教育部机器智能与先进计算重点实验室）
DGNS: Deformable Gaussian Splatting and Dynamic Neural Surface for Monocular Dynamic 3D Reconstruction
Industrial Research Organisation, Canberra, ACT Australia；College of Engineering Computing \& Cybernetics, The Australian National University, Canberra, ACT Australia；Faculty of Engineering, The University of Hong Kong, Hong Kong SAR China；Industrial Research Organisation；College of Engineering Computing \& Cybernetics, The Australian National University；Faculty of Engineering, The University of Hong Kong
${C}^{3}$-GS: Learning Context-aware, Cross-dimension, Cross-scale Feature for Generalizable Gaussian Splatting
Institute of Visual Computing, Graz University of Technology（视觉计算研究所，格拉茨技术大学）；School of Electronic and Computer Engineering, Peking University（电子与计算机工程学院，北京大学）
DET-GS: Depth- and Edge-Aware Regularization for High-Fidelity 3D Gaussian Splatting
Perceptual Imaging Laboratory (PILab), School of Electrical and Data Engineering, Faculty of Engineering and Information Technology, University of Technology Sydney（感知成像实验室（PILab）、电气与数据工程学院、工程与信息技术学院、新南威尔士大学）
Learning Fine-Grained Geometry for Sparse-View Splatting via Cascade Depth Loss
School of Computer Science The University of Sydney Sydney, Australia；School of Computer Science University of Bristol Bristol, United Kingdom；School of Science Edith Cowan University Perth, Australia
X-Splat: Gaussian Splatting for 3D CBCT Generation from Single Panoramic Radiograph
Sano Centre for Computational Medicine（桑诺计算医学中心）；Jagiellonian University（雅盖隆大学）；Warsaw University of Technology（华沙技术大学）；Research Institute IDEAS（IDEAS研究 institute）；Harvard Medical School（哈佛医学院）
