English(EN)Learning Implicit Constitutive Laws for Dynamic 3D Gaussian Splatting from Monocular Videos
3D高斯溅射研究在版权、压缩和动态场景建模方面取得进展 · 已追踪10个来源
作者PulseAugur 编辑部·[77 个来源]·
近期研究通过各种创新推动了3D高斯溅射（3DGS），包括用于版权保护的X-SG$^2$S方法、用于改进重建质量的DualDiff3D以及用于理解表面恢复的理论框架。进一步的开发通过语义运动图（SMG）和高斯构成对齐（GCA）解决了动态场景问题，而像KISS-GS这样的压缩技术旨在减小巨大的文件大小。还出现了专业应用，例如AquaFlow和NemoSplat的水下重建，以及CC-4DGS的存储高效动态4DGS。
AI
Recent single-view feed-forward 3D Gaussian Splatting (3DGS) generation predicts a fixed number of Gaussians per camera ray, introducing severe spatial redundancy. Most existing compaction strategies target multi-view setups to exploit cross-view consistency and are incompatible …
Conventional 3D Gaussian Splatting assumes a closed set of observations and long optimization schedules. Continual RGB-D mapping in contrast poses the problem that new observations arrive online, while previously reconstructed regions must be preserved. We present EliGSiR (Eviden…
arXiv:2609.11434v1 Announce Type: cross Abstract: We introduce Complex-Valued Quadratic Phase Gaussian (CVQPG), a novel hologram representation method that replaces standard 2D Gaussian representations used in 2D Gaussian Splatting with 2D quadratic phase functions. CVQPG incorpo…
Recovering clean 3D scenes from hazy multi-view images is challenging because haze attenuates scene radiance and introduces atmospheric scattering. Recent scattering-aware Gaussian Splatting methods introduce physical haze models into reconstruction, but they often apply degradat…
arXiv:2609.05925v1 Announce Type: cross Abstract: Pose-free feed-forward 3D Gaussian Splatting enables novel view synthesis from uncalibrated multi-view images. Although more views should improve performance, existing methods often degrade with dense-view inputs because global ag…
Sparse-view X-ray 3D reconstruction is essential for reducing radiation exposure, but recovering a density field from a handful of X-ray projections is severely ill-posed. Recently, 3D Gaussian Splatting has achieved state-of-the-art performance in sparse-view reconstruction by r…
3D Gaussian Splatting (3DGS) renders novel views in real time, but an uncertainty heatmap does not certify that a rendered view meets a certain prediction coverage. We treat novel-view synthesis as structured regression and ask that, with probability at least $1-α$, RGB predictio…
arXiv cs.AI
TIER_1English(EN)·Theo Morales, Nhat-Quynh Le-Pham, Robin Atkins, Binh-Son Hua·
arXiv:2609.03534v1 Announce Type: cross Abstract: 3D Gaussian Splatting has become a de facto scene representation for novel view synthesis, yet robustly learning 3D Gaussian primitives from visual input remains challenging. Standard optimization relies on gradient-based updates,…
3D Gaussian Splatting has become a de facto scene representation for novel view synthesis, yet robustly learning 3D Gaussian primitives from visual input remains challenging. Standard optimization relies on gradient-based updates, but a common issue is the gradient vanishing phen…
Achieving truly immersive large-scale scene digitization necessitates consistent and visually pleasing rendering across all possible viewing perspectives. However, collecting multi-view images covering every fine detail of a large-scale scene is prohibitive due to scene complexit…
arXiv cs.AI
TIER_1English(EN)·Zihang Cheng, Wentao Bao, Huiping Zhuang, Chun Li, Xin Meng, Ziqian Zeng, Cen Chen, Ming Li, F. Richard Yu·
arXiv:2502.10475v3 Announce Type: replace-cross Abstract: 3D Gaussian Splatting (3DGS) has been widely used in 3D reconstruction and 3D generation. However, the rapid adoption of 3D Gaussian Splatting raises growing concerns about information leakage and unauthorized use, urging …
While 3D Gaussian Splatting (3DGS) has revolutionized 3D reconstruction and novel-view synthesis, scenarios with limited input views often lead to poor reconstruction quality and artifacts in rendered novel views. Recent efforts attempt to utilize powerful diffusion priors, yet t…
arXiv cs.LG
TIER_1English(EN)·Songhe Wang, David Johnathan Miller·
arXiv:2608.30054v1 Announce Type: new Abstract: When does 3D Gaussian Splatting (3DGS) recover the true scene surface rather than just overfitting view-dependent appearance? We answer this by developing a mathematical framework based on a first-hit rendering abstraction that clea…
We study dynamic Gaussian Splatting from monocular videos. While recent advancements in dynamic Gaussian splatting offer a promising foundation for modeling dynamic scenes, they often overfit to the training views and fail under occlusion or complex scene motion due to the lack o…
Splatting-based algorithms reconstruct photorealistic, real-time-renderable, and mesh-exportable 3D scenes from regular images, but they represent a scene as a single monolithic field. Therefore, the reconstruction has no object-level structure, leaving it infeasible for downstre…
Scene reconstruction with 3D Gaussian Splatting (3DGS) has become common, however deployment remains painful as the uncompressed file sizes can be massive. Current 3DGS compression systems combine multiple strategies for file size reduction, which can obscure where gains come fro…
The underwater environment is challenging for 3D reconstruction, because particles suspended in the water scatter and diffuse light, turbidity varies, absorption depends on wavelength, and illumination is rarely uniform. Methods based on Gaussian splatting have generally been dev…
arXiv cs.AI
TIER_1English(EN)·Xiaoyang Liu, Kai Han·
arXiv:2608.22102v1 Announce Type: cross Abstract: We present GCA (Gaussian Constitutive Alignment), a framework for learning implicit constitutive laws from monocular dynamic video of deformable objects represented by 3D Gaussians. Given a static multi-view scan for geometric ini…
Recent monocular 3D Gaussian Splatting (3DGS) streaming reconstruction methods have achieved impressive performance by balancing reconstruction quality and efficiency. However, extending these frameworks to underwater scenes remains challenging due to severe visual degradation, s…
Reconstructing photorealistic scenes in unconstrained underwater environments remains challenging due to severe media-induced light scattering and unpredictable dynamic objects. Recent feed-forward visual foundation models have demonstrated remarkable capabilities in generalized …
arXiv cs.CV
TIER_1English(EN)·Bj\"orn Ellensohn, Elmar Rueckert, Christian Rauch·
arXiv:2609.20348v1 Announce Type: new Abstract: Conventional 3D Gaussian Splatting assumes a closed set of observations and long optimization schedules. Continual RGB-D mapping in contrast poses the problem that new observations arrive online, while previously reconstructed regio…
arXiv:2609.19628v1 Announce Type: new Abstract: We present VGGT-GS SLAM, a monocular 3D Gaussian Splatting SLAM system designed for uncalibrated videos. Starting from feed-forward VGGT pose and depth priors, our system performs submap differentiable bundle adjustment that jointly…
arXiv cs.CV
TIER_1English(EN)·Xiaobiao Du, Yida Wang, Cheng Bi, Kun Zhan, Xin Yu·
arXiv:2609.19421v1 Announce Type: new Abstract: Gaussian Splatting has significantly improved the quality of novel view synthesis with explicit Gaussian representation. However, we observed that existing 3D Gaussian Splatting methods (3DGS) often suffer from surface collapse issu…
arXiv cs.CV
TIER_1English(EN)·Zhikun Zhou, Kunyu Peng, Runyi Yang, Junhao Cai, Di Wen, Ruiping Liu, Danda Pani Paudel, Yi Zhou, Luc Van Gool, Kailun Yang·
arXiv:2609.20586v1 Announce Type: cross Abstract: Referring scene understanding for embodied robots requires grounding object- and relation-centric language queries from a designated viewpoint. While a local semantic Gaussian map can support such grounding within one agent's obse…
arXiv:2609.20623v1 Announce Type: new Abstract: Recent single-view feed-forward 3D Gaussian Splatting (3DGS) generation predicts a fixed number of Gaussians per camera ray, introducing severe spatial redundancy. Most existing compaction strategies target multi-view setups to expl…
arXiv cs.CV
TIER_1English(EN)·Kristof Overdulve, Lode Jorissen, Nick Michiels·
arXiv:2609.18473v1 Announce Type: new Abstract: We present CADSplat, a framework that reconstructs photorealistic, geometrically accurate digital twins from sparse ($<15$ views), wide-baseline posed images of an object by regularizing 3D Gaussian Splatting (3DGS) with an explicit…
arXiv:2609.18737v1 Announce Type: new Abstract: Underwater 3D reconstruction supports applications ranging from marine ecosystem monitoring and subsea inspection to underwater archaeology, education, and immersive visualisation. 3D Gaussian Splatting has made real-time photoreali…
arXiv cs.CV
TIER_1English(EN)·Yihan Zang, Da Li, Dominik Engel, Shinkyu Park, Ivan Viola·
arXiv:2609.18898v1 Announce Type: new Abstract: Training-free weighted aggregation is widely used to lift 2D semantic features onto 3D Gaussians for open-vocabulary scene understanding, yet its theoretical role remains insufficiently understood. Existing analyses typically justif…
arXiv:2511.21265v2 Announce Type: replace Abstract: Learning-based image matching critically depends on large-scale, diverse, and geometrically accurate training data. 3D Gaussian Splatting (3DGS) enables photorealistic novel-view synthesis and thus is attractive for data generat…
arXiv:2609.17387v1 Announce Type: new Abstract: Real-time dense SLAM is a core capability for robotics applications that require robust localization and high- quality mapping in dynamic or fast-changing environments. Recent 3D Gaussian Splatting (3DGS)-based SLAM methods have sho…
arXiv cs.CV
TIER_1English(EN)·Tam Thuc Do, Philip A. Chou, Gene Cheung·
arXiv:2609.15735v1 Announce Type: cross Abstract: We address the problem of color attribute compression for 3D splats. We show that all images generated by 3D splats are linear in the coefficients for each color channel, each spherical harmonic, and each splat, and we identify a …
arXiv:2609.14129v1 Announce Type: cross Abstract: Ultra-High-Definition (UHD) video presents significant challenges for efficient storage and real-time decoding. Learning-based methods, such as Neural Video Compression (NVC) and Implicit Neural Representations (INR), achieve comp…
arXiv:2609.13262v1 Announce Type: new Abstract: 3D Gaussian Splatting achieves photorealistic reconstruction within training view distribution, yet it degrades on out-of-distribution novel views, exhibiting holes in unobserved regions and artifacts in observable areas. Recent wor…
arXiv:2609.12682v1 Announce Type: new Abstract: Reconstructing 3D scenes under real-world low-light conditions remains challenging due to severe sensor noise, low signal-to-noise ratios, and degraded photometric consistency, which destabilize geometry estimation and novel view sy…
arXiv:2609.12343v1 Announce Type: new Abstract: Feed-forward Gaussian splatting models have demonstrated remarkable effectiveness in reconstructing three-dimensional (3D) objects from a few two-dimensional (2D) images, even if they are unseen. As existing methods typically predic…
arXiv:2512.07197v2 Announce Type: replace Abstract: 3D Gaussian Splatting (3DGS) has emerged as a powerful explicit representation enabling real-time, high-fidelity 3D reconstruction and novel view synthesis. However, its practical use is hindered by the massive memory and comput…
arXiv cs.CV
TIER_1English(EN)·Dingqiang Ye, Jiacong Xu, Jianglu Ping, Yuxiang Guo, Chao Fan, Vishal M. Patel·
arXiv:2603.11298v3 Announce Type: replace Abstract: High dynamic range (HDR) novel view synthesis (NVS) aims to reconstruct HDR scenes from multi-exposure low dynamic range (LDR) images. Existing HDR pipelines heavily rely on known camera poses, well-initialized dense point cloud…
arXiv:2609.11223v1 Announce Type: new Abstract: Recovering clean 3D scenes from hazy multi-view images is challenging because haze attenuates scene radiance and introduces atmospheric scattering. Recent scattering-aware Gaussian Splatting methods introduce physical haze models in…
arXiv:2609.06017v2 Announce Type: replace Abstract: The appearance of a smoky scene is shaped by two processes that a camera records together: the participating medium alters scene radiance in a view-dependent way, and the image signal processor (ISP) then remaps the result throu…
arXiv cs.CV
TIER_1English(EN)·Junzheng Chu, Bin Pan, Zhenwei Shi·
arXiv:2609.10307v1 Announce Type: cross Abstract: 3D Gaussian Splatting (3DGS) renders novel views in real time, but an uncertainty heatmap does not certify that a rendered view meets a certain prediction coverage. We treat novel-view synthesis as structured regression and ask th…
arXiv:2609.10095v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) enables real-time novel view synthesis but produces millions of primitives through adaptive densification, leading to significant storage overhead. Learned-mask pruning methods such as LP-3DGS address th…
arXiv:2609.10376v1 Announce Type: new Abstract: Sparse-view X-ray 3D reconstruction is essential for reducing radiation exposure, but recovering a density field from a handful of X-ray projections is severely ill-posed. Recently, 3D Gaussian Splatting has achieved state-of-the-ar…
arXiv cs.CV
TIER_1English(EN)·Florian Hahlbohm, Jorge Condor, Linus Franke, Martin Eisemann, Marcus Magnor·
arXiv:2609.05255v1 Announce Type: new Abstract: Explicit primitive-based radiance fields such as 3D Gaussian Splatting typically model view-dependent appearance using low-order spherical harmonics (SH). Although efficient to evaluate, SH coefficients dominate per-primitive storag…
arXiv:2609.03334v1 Announce Type: new Abstract: A key bottleneck in 3D Gaussian Splatting training is the continual growth of Gaussian primitives, which increases optimization cost and slows convergence, especially at high resolutions. We propose Laplacian Frequency Hierarchies, …
arXiv cs.CV
TIER_1English(EN)·Bocheng Li, Wenjuan Zhang, Jie Pan. Dongxu Han, Xuesong Ma, Yiling Yao, Yaning Wang·
arXiv:2609.03447v1 Announce Type: new Abstract: Large-scale 3D surface reconstruction from aerial imagery is fundamental to geospatial mapping and urban modeling. Recent advances in 3D Gaussian Splatting (3DGS) have demonstrated considerable potential for this task. However, exis…
arXiv:2509.15677v2 Announce Type: replace Abstract: We propose Camera Splatting, a novel view optimization framework for novel view synthesis. Each camera is modeled as a 3D Gaussian, referred to as a camera splat, and virtual cameras, termed point cameras, are placed at 3D point…
arXiv:2603.09968v2 Announce Type: replace Abstract: Online novel view synthesis requires a model to reconstruct a scene causally from a stream of observations while keeping it renderable at every moment. We present ReCoSplat, an online feed-forward Gaussian Splatting model suppor…
arXiv cs.CV
TIER_1English(EN)·Injae Kim, Chaehyeon Kim, Minseong Bae, Minseok Joo, Hyunwoo J. Kim·
arXiv:2603.21304v3 Announce Type: replace Abstract: Feed-forward 3D Gaussian Splatting methods enable single-pass reconstruction and real-time rendering. However, they typically adopt rigid pixel-to-Gaussian or voxel-to-Gaussian pipelines that uniformly allocate Gaussians, leadin…
arXiv cs.CV
TIER_1English(EN)·Kyungdae Park, Chae Eun Rhee·
arXiv:2609.02184v1 Announce Type: new Abstract: Dynamic four-dimensional (4D) Gaussian Splatting has emerged as a powerful explicit representation for high-quality view synthesis, yet existing methods still require tens to hundreds of megabytes per scene due to their heavy relian…
arXiv cs.CV
TIER_1English(EN)·Wentao Sun, Yiping Chen, John S. Zelek, Jonathan Li·
arXiv:2508.00259v2 Announce Type: replace Abstract: While 3D Gaussian Splatting (3DGS) has established new standards for high-fidelity 3D scene modeling, interpreting massive, unstructured Gaussian primitives into meaningful geospatial entities remains a critical challenge for re…
arXiv cs.CV
TIER_1English(EN)·Tianheng Lu, Guangyu Wang, Ruqi Huang, Lu Fang·
arXiv:2609.02747v1 Announce Type: new Abstract: Achieving truly immersive large-scale scene digitization necessitates consistent and visually pleasing rendering across all possible viewing perspectives. However, collecting multi-view images covering every fine detail of a large-s…
arXiv cs.CV
TIER_1English(EN)·Wei Dong, Shahram Shirani, Jun Chen, Han Zhou·
arXiv:2609.00994v1 Announce Type: new Abstract: Recent extensions of 3D Gaussian Splatting (3DGS) enable real-time novel view synthesis in dynamic scenes by learning time-conditioned Gaussian deformations. However, existing MLP-based methods typically estimate deformations indepe…
arXiv:2609.01516v1 Announce Type: new Abstract: While 3D Gaussian Splatting (3DGS) has revolutionized 3D reconstruction and novel-view synthesis, scenarios with limited input views often lead to poor reconstruction quality and artifacts in rendered novel views. Recent efforts att…
arXiv cs.CV
TIER_1English(EN)·Haozheng Yu, Xinyu Yang, Rundong Luo, Jennifer J. Sun, Bharath Hariharan·
arXiv:2608.31023v1 Announce Type: new Abstract: We study dynamic Gaussian Splatting from monocular videos. While recent advancements in dynamic Gaussian splatting offer a promising foundation for modeling dynamic scenes, they often overfit to the training views and fail under occ…
arXiv cs.CV
TIER_1English(EN)·Jiahao Wu, Jie Liang, Die Hu, Jiayu Yang, Kaiqiang Xiong, Xiang Li, Xiaoyun Zheng, Chao Wang, Ronggang Wang·
arXiv:2608.30184v1 Announce Type: new Abstract: Volumetric video enables immersive free viewpoint rendering of dynamic real world scenes, yet existing methods struggle with long sequences and complex motions, often leading to temporal instability and visual artifacts. To address …
arXiv:2608.29346v1 Announce Type: new Abstract: 3D Gaussian Splatting has achieved remarkable success in photorealistic rendering, yet it suffers from severe overfitting and geometric artifacts in sparse-view scenarios due to the inherent deficiency of photometric supervision. Re…
arXiv cs.CV
TIER_1English(EN)·Zhongliang Liu, Wenjie Liu, Yang Li·
arXiv:2608.29048v1 Announce Type: new Abstract: Reference-guided stylization of scenes represented by 3D Gaussian Splatting (3DGS) is important for efficient and controllable 3D content creation. Existing VGG-feature-based 3D stylization methods provide stable rendered-view optim…
arXiv:2608.28702v1 Announce Type: new Abstract: Deformable 3D Gaussian Splatting (D-3DGS) reconstructs dynamic scenes by deforming a canonical set of 3D Gaussians through a deformation field of frame time. Replacing its MLP with a stack of Closed-form Continuous-time (CfC) cells-…
arXiv:2603.09703v2 Announce Type: replace Abstract: Progressive transmission of 3D Gaussian Splatting (3DGS) requires each completed transmission stage to be decodable from received data and directly renderable. This work presents ProGS, a progressive codec that organizes anchor-…
arXiv cs.CV
TIER_1English(EN)·Zhaoqi Su, Shihai Chen, Xinyan Lin, Liqin Huang, Zhipeng Su, Xiaoqiang Lu·
arXiv:2601.15897v3 Announce Type: replace Abstract: Multi-modal scene reconstruction integrating RGB and thermal infrared data is essential for robust environmental perception across diverse lighting and weather conditions. However, extending 3D Gaussian Splatting (3DGS) to multi…
arXiv cs.CV
TIER_1English(EN)·Hannah Schieber, Dominik Frischmann, Victor Schaack, Simon Boche, Angela Schoellig, Stefan Leutenegger, Daniel Roth·
arXiv:2509.04859v4 Announce Type: replace Abstract: Fast and efficient photorealistic 3D reconstruction with (semantic) Gaussian Splatting (GS) is crucial for time-critical robotic perception and navigation, where robots may need to rapidly reconstruct and inspect specific points…
arXiv:2406.00434v4 Announce Type: replace Abstract: In this paper, we propose MoDGS, a new pipeline to render novel views of dy namic scenes from a casually captured monocular video. Previous monocular dynamic NeRF or Gaussian Splatting methods strongly rely on the rapid move men…
arXiv:2608.28240v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) delivers real-time and high-fidelity rendering but remains challenged by unconstrained in-the-wild scenes, where drastic appearance variations and transient objects violate multi-view consistency. Existi…
arXiv cs.CV
TIER_1English(EN)·Haofei Xu, Daniel Barath, Andreas Geiger, Marc Pollefeys·
arXiv:2510.08575v4 Announce Type: replace Abstract: While existing feed-forward Gaussian splatting models offer computational efficiency and can generalize to sparse view settings, their performance is fundamentally constrained by relying on a single forward pass for inference. W…
arXiv cs.CV
TIER_1English(EN)·Wieland Morgenstern, Friedrich Elias Branschke, Florian Fleischmann, Adrian Szatmari, Paul Schlack, Florian Barthel, Peter Eisert, Anna Hilsmann·
arXiv:2608.26948v1 Announce Type: new Abstract: Scene reconstruction with 3D Gaussian Splatting (3DGS) has become common, however deployment remains painful as the uncompressed file sizes can be massive. Current 3DGS compression systems combine multiple strategies for file size r…
arXiv cs.CV
TIER_1English(EN)·Jean-Daniel de Ambrogi, Aladine Chetouani, Vincent Nguyen, Aur\'elien Chateigner·
arXiv:2608.26868v1 Announce Type: new Abstract: Recent advances in SLAM have leveraged 3DGS for photorealistic reconstruction and novel view synthesis. However, most methods rely on RGB-D input, which is unavailable on consumer-grade smartphones, and few integrate 3DGS within a c…
arXiv:2603.07587v2 Announce Type: replace Abstract: 3D Gaussian Splatting (3DGS) has demonstrated remarkable performance in novel view synthesis and 3D scene reconstruction, but its quality often degrades in real-world environments due to transient distractors, such as moving obj…
arXiv:2608.25483v1 Announce Type: new Abstract: The underwater environment is challenging for 3D reconstruction, because particles suspended in the water scatter and diffuse light, turbidity varies, absorption depends on wavelength, and illumination is rarely uniform. Methods bas…
arXiv:2608.21849v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) has achieved remarkable success in novel view synthesis; however, reconstructions under sparse views often exhibit noticeable artifacts. While recent video diffusion models provide strong spatio-temporal…
arXiv cs.CV
TIER_1English(EN)·Pawel Tomasz Pieta, Rasmus Juul Pedersen, Sina Borgi, Jakob Sauer J{\o}rgensen, Jens Wenzel Andreasen, Vedrana Andersen Dahl·
arXiv:2604.01844v2 Announce Type: replace Abstract: Gaussian Splatting (GS) has emerged as a dominating technique for image rendering and has quickly been adapted for the X-ray Computed Tomography (CT) reconstruction task. However, despite its growing popularity, the benefits of …
arXiv:2608.22906v1 Announce Type: new Abstract: Recent monocular 3D Gaussian Splatting (3DGS) streaming reconstruction methods have achieved impressive performance by balancing reconstruction quality and efficiency. However, extending these frameworks to underwater scenes remains…
arXiv:2608.22888v1 Announce Type: new Abstract: Reconstructing photorealistic scenes in unconstrained underwater environments remains challenging due to severe media-induced light scattering and unpredictable dynamic objects. Recent feed-forward visual foundation models have demo…
arXiv:2608.22773v1 Announce Type: new Abstract: Dynamic 3D Gaussian Splatting (3DGS) achieves photorealistic reconstruction of time-varying scenes, and recent physics-aware extensions improve extrapolation by explicitly predicting velocity fields. However, these extensions merely…
arXiv:2508.03077v2 Announce Type: replace Abstract: Feedforward 3D Gaussian Splatting (3DGS) overcomes the limitations of optimization-based 3DGS by enabling fast and high-quality reconstruction without the need for per-scene optimization. However, existing feedforward approaches…
arXiv:2608.22344v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) achieves state-of-the-art rendering quality at real-time speeds but suffers from "model bloat" - a large number of redundant, low-opacity Gaussians that inflate memory usage and training costs. This inef…
