GrainGS is a dynamic Gaussian framework that combines a hierarchical anchor scaffold with per-Gaussian deformation that achieves high reconstruction quality, real-time novel view synthesis, and compact storage.
Abstract
Dynamic scene reconstruction with 3D Gaussian Splatting requires a balance between fine-grained motion modeling, structural stability, and compact representation. Existing per-primitive methods provide flexible local deformation but often suffer from redundant primitive growth, while anchor-based methods improve spatial regularity at the cost of suppressing locally varying motion. To address these issues, we present GrainGS, a dynamic Gaussian framework that combines a hierarchical anchor scaffold with per-Gaussian deformation. A static warm-up stage first establishes a time-invariant canonical representation from observations across all timestamps. During joint training, a stop-gradient operation blocks the deformation-mediated gradient pathway to the canonical positions while preserving their direct refinement through the reconstruction objective. Each Gaussian then predicts independent temporal offsets for position, rotation, and scale, enabling detailed local motion within a structurally constrained scaffold. A canonical-residual appearance decomposition further models frame-dependent photometric changes without forcing them into geometric deformation. Experiments on synthetic monocular and real-world multiview benchmarks show that GrainGS achieves high reconstruction quality, real-time novel view synthesis, and compact storage. Under the synthetic benchmark setting, it reaches an average peak signal-to-noise ratio of 36.98 decibels, renders at 435.6 frames per second, and requires 4.67 megabytes of storage.
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 phenomenon: a pixel far from a Gaussian primitive often has diminishing gradient magnitudes to influence primitive attributes, resulting in suboptimal scene reconstruction. In this paper, we propose a method to address gradient vanishing with a piecewise truncated gradient formulation that improves the optimization stability and robustness to initializations. We show that our method consistently improves 3D Gaussian Splatting with random and COLMAP initializations while being generalizable across static and dynamic Gaussian Splatting. As a by-product, we also examine the limitations of current benchmarks for dynamic scenes, and introduce a novel dataset for benchmarking dynamic Gaussian Splatting using synthetic 3D scenes. We demonstrate the effectiveness of our method in both static and dynamic settings for the public benchmarks and our proposed dataset.
Theo Morales, Nhat-Quynh Le-Pham, Robin Atkins et al.· 0 citations
TaylorMoDe-GS, the first 3DGS framework tailored for multi-view dynamic object deblurring, shifts the modeling paradigm from displacement fitting to velocity driven modeling, and introduces a neural Peano remainder network to compensate for high frequency non-linear dynamics.
Xiaofeng Quan, Junzhe Wan, Chao Cai et al.· 0 citations
Editing dynamic scenes with 4D Gaussian Splatting (4DGS) is often hampered by spatiotemporal inconsistencies, or "Gaussian drifting", which degrades edit quality and temporal coherence. We identify that these artifacts stem from two distinct sources: foundational inaccuracies in the initial scene reconstruction, and the disruption of learned trajectories during the editing process itself. To address this, we propose a comprehensive framework that systematically tackles both sources of inconsistency. To solve reconstruction-induced errors, we introduce a novel prior-guided, multi-stage reconstruction pipeline that fuses geometric and motion priors to build a physically plausible and temporally stable foundation. To solve editing-induced errors, we further apply a universal trajectory-preserving technique, which safeguards high-quality motion by decoupling the appearance optimization from the learned deformation. Experiments demonstrate that by systematically addressing both the reconstruction and editing phases, our method achieves state-of-the-art, temporally consistent editing on a wide range of dynamic scenes where previous monolithic approaches fail.
Xiaosheng He, Feng-Lin Liu, Lin Gao et al.· IEEE Transactions on Visuali...· 0 citations
Existing 3D mesh reconstruction methods from Gaussian scene representations predominantly rely on iterative optimization, resulting in slow inference and limited scalability to high-resolution inputs. In this paper, we present AnyGS2Mesh, the first feed-forward framework for directly reconstructing 3D meshes from 3D Gaussian Splatting representations with support for arbitrary input image resolutions. Our approach incorporates a Gaussian-Guided Transformer architecture that exploits explicit 3D geometric priors for efficient mesh generation. We introduce three key components: (1) a Gaussian-Guided Spatial Reasoning Transformer represents Gaussian primitives as structured 3D tokens and jointly reasons over Gaussian and image features; (2) a Streaming and Patchwise Geometry Encoder processes native-resolution views sequentially and aggregates information across variable-length view sets; (3) a Scale-Aligned Hybrid Depth Refiner uses a PatchFusion-style encoder--decoder to fuse RGB-conditioned predicted depth with Gaussian-rendered metric depth, combining fine local structures with globally consistent metric scale. The refined depth maps are integrated through TSDF fusion, followed by Marching Cubes for deterministic mesh extraction. Extensive experiments show that AnyGS2Mesh achieves state-of-the-art reconstruction quality while significantly reducing inference time compared with optimization-based baselines, enabling near-real-time, high-quality mesh reconstruction. Our results demonstrate the potential of combining Gaussian representations and feed-forward Transformer architectures for scalable 3D geometry reconstruction. The code will be made publicly available upon acceptance.
Yuxuan Song, Fan Gao, Yi-Bo Zhao et al.· 0 citations
Maintaining global geometric consistency is a central challenge in long-sequence 3D reconstruction, with scale drift being the most critical failure mode. In chunk-based inference pipelines, the scale degree of freedom in sequential Sim(3) alignment is left unconstrained, causing estimation errors to compound multiplicatively and distort global trajectories and point cloud geometry. We present a scale-consistency enhancement framework built on a key insight: in structured environments such as driving scenes, geometric quantities arising from environmental regularity remain inherently invariant across temporal segments, and discrepancies in their per-chunk measurements directly expose inter-chunk scale drift. We propose Scene Geometric Invariant Anchoring (SGIA), which extracts dominant geometric invariants from each chunk's predicted point cloud via coarse-to-fine robust estimation and exploits their cross-chunk consistency to establish scale constraints independent of point cloud registration, explicitly degenerating 7-DoF Sim(3) alignment into 6-DoF rigid-body transformation and severing chain-wise scale error propagation at its source. We further introduce a lightweight test-time adaptation strategy that fine-tunes only normalization-layer parameters via multi-objective self-supervision, progressively improving intra-chunk predictions along the sequence. Both modules are plug-and-play and require no offline retraining. Experiments on multiple long-sequence benchmarks demonstrate state-of-the-art performance, reducing absolute trajectory error by up to 32% with significant gains in trajectory stability and reconstruction quality. Code: https://github.com/WZ-CS/VGGT-Align
This paper proposes a semantics-guided scene decoupling module that separates Gaussian primitives into static and dynamic components based on motion vectors, and introduces a motion-aware densification module for motion compensation, which alleviates the incomplete rendering of dynamic objects caused by insufficient spatio-temporal information.
Chulin Zhao, Xue Wang, Guoqing Zhou et al.· IEEE Transactions on Visuali...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
