English(EN)GenSplatCodec: Feed-Forward Gaussian Splatting Compression via One-Step Diffusion
新的3D高斯喷涂方法增强场景编辑、重建和动画
作者PulseAugur 编辑部·[51 个来源]·
近期研究探索了用于场景编辑、重建和动画的3D高斯喷涂（3DGS）的进展。LB-Edit引入了注意力引导的相机放置和多视图对齐，以实现更精确的3D场景编辑。ATSplat通过使用自适应3D标记来增强前馈3DGS，以改进场景复杂性表示并减少高斯数量。ZeroSplat将3DGS中的引用分割泛化到处理多个目标，并通过将2D视觉语言模型先验提升到3D空间来提高效率。FF-ProCams为投影仪-相机系统提供了一个前馈框架，实现了快速逆渲染。E3DGS通过将颜色视为几何实体来统一几何-光度等变性，提高了鲁棒性和数据效率。AsySplat提出了一种用于高效动态场景建模的不对称架构，将几何和外观解耦。CGGS专注于以一致性增强损失和改进的几何重建进行以自我为中心的3D场景生成。3D-GIMP将扩散模型与PatchMatch结合，用于3DGS中的高保真对象移除，优先考虑一致性和速度。GrainGS通过结合分层锚点支架与每个高斯的变形和梯度解耦来解决动态场景重建问题，以实现高效的运动建模。最后，HumanGS提供了一个用于从稀疏多视图图像和姿势数据进行实时人类重建和动画的前馈框架。
AI
Reconstructing articulated objects with multiple movable parts is essential for understanding object structure and enabling physical interaction. However, this reconstruction task poses significant challenges due to the entanglement of geometry, appearance, and motion parameters …
3D Gaussian Splatting (3DGS) has emerged as an effective representation for novel view synthesis and 3D scene reconstruction, creating an increasing demand for reliable quality assessment. Unlike conventional image quality assessment (IQA), the quality of a 3DGS scene depends not…
Feed-forward 3D Gaussian Splatting (3DGS) enables scalable scene reconstruction without per-scene optimization, yet produces dense Gaussians that are costly to store and transmit. Existing feed-forward Gaussian compression methods formulate decoding as deterministic representatio…
Text-driven 3D scene editing with 3D Gaussian Splatting (3DGS) typically applies a 2D diffusion editor to views rendered from fixed training cameras, limiting both the spatial coverage of edits and the user's freedom to target specific objects in complex scenes. We present LB-Edi…
3D Gaussian Splatting (3DGS) achieves high-quality novel-view synthesis by optimizing freely placed primitives in 3D and adaptively densifying them in under-reconstructed regions. However, this scene-adaptive capacity allocation is largely lost in existing feed-forward 3DGS metho…
Recent advancements in 3D Gaussian Splatting (3DGS) have enabled language-guided scene understanding. However, existing Referring 3D Gaussian Splatting (R3DGS) methods are fundamentally restricted to single-target queries. To reflect the ambiguity of real-world instructions, we i…
3D Gaussian Splatting (3DGS) employs Gaussian primitives for explicit scene representation, facilitating real-time, high-fidelity reconstruction and novel view synthesis of complex scenes. However, the explicit modeling inherent in 3DGS introduces a gradient bias during optimizat…
Projector-camera (ProCams) systems achieve active scene perception and controllable appearance manipulation via structured illumination, serving as a core infrastructure for spatial augmented reality, projection mapping, and surface reflectance acquisition. Existing inverse-rende…
arXiv cs.LG
TIER_1English(EN)·Chankyo Kim, Maani Ghaffari·
arXiv:2607.15536v1 Announce Type: cross Abstract: 3D Gaussian Splatting (3DGS) captures scenes by coupling explicit geometry (position, covariance) with view-dependent photometry (Spherical Harmonics). However, building $\mathrm{SE}(3)$-equivariant architectures on these primitiv…
Recent generalizable 3D Gaussian Splatting models have advanced long-sequence novel view synthesis (NVS), but at the cost of substantial redundant computation. We identify that the redundancy can be mitigated based on two observations: (i) high-precision geometry is not strictly …
Challenges remain in ego-centric 3D scene generation due to limited view overlap and the dominant influence of individual perspectives on scene interpretation. These factors hinder the creation of viewpoint-consistent and semantically aligned visual content, as well as the constr…
arXiv:2607.26595v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) has emerged as an effective representation for novel view synthesis and 3D scene reconstruction, creating an increasing demand for reliable quality assessment. Unlike conventional image quality assessmen…
arXiv:2607.26889v1 Announce Type: cross Abstract: Reconstructing articulated objects with multiple movable parts is essential for understanding object structure and enabling physical interaction. However, this reconstruction task poses significant challenges due to the entangleme…
arXiv cs.CV
TIER_1English(EN)·Tianyi Song, Danail Stoyanov, Evangelos Mazomenos, Francisco Vasconcelos·
arXiv:2508.13287v3 Announce Type: replace-cross Abstract: 3D Gaussian Splatting (3DGS) has recently gained popularity for efficient scene rendering by representing scenes as explicit sets of anisotropic 3D Gaussians. However, most existing work focuses primarily on modeling exter…
arXiv:2607.22780v1 Announce Type: cross Abstract: Faithful inverse rendering requires visibility and indirect radiance to explain secondary illumination and inter-reflection, yet rasterization-oriented Gaussian representations do not naturally support the secondary-ray queries ne…
arXiv:2503.14736v3 Announce Type: replace Abstract: Photorealistic and animatable hand avatars are essential for applications such as AR/VR, gaming, and telepresence. Recent 3D Gaussian Splatting (3DGS) based avatar methods enable real-time rendering of articulated humans, but mo…
arXiv cs.CV
TIER_1English(EN)·Devdoot Chatterjee, Zakaria Laskar, C. V. Jawahar·
arXiv:2604.10259v2 Announce Type: replace Abstract: We present HumanGS, a generalizable feed-forward Gaussian splatting framework for human 3D reconstruction and real-time animation from sparse multi-view RGB images and their associated SMPL-X poses. Unlike prior methods that rel…
arXiv:2607.20789v1 Announce Type: new Abstract: Recent advances in 3D scene editing have leveraged iterative diffusion models to update input views. However, this process is computationally expensive and struggles to produce sharp details. Meanwhile, ``hallucination drift'' frequ…
arXiv cs.CV
TIER_1English(EN)·Jiahao He, Yihua Shao, Zhengkai Zhao, Pan Gao, Fei Ma, Jingcai Guo, Hao Tang, Nicu Sebe, Qi Tian·
arXiv:2607.21448v1 Announce Type: new 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…
arXiv cs.CV
TIER_1English(EN)·Rong Fu, Jiekai Wu, Haiyun Wei, Yee Tan Jia, Yang Li, Xiaowen Ma, Wangyu Wu, Simon Fong·
arXiv:2602.15355v4 Announce Type: replace Abstract: The emergence of 3D Gaussian Splatting has fundamentally redefined the capabilities of photorealistic neural rendering by enabling high-throughput synthesis of complex environments. While procedural methods like Wang Tiles have …
arXiv cs.CV
TIER_1English(EN)·Jaeyeon Park, Taeho Kang, Youngki Lee·
arXiv:2607.19777v1 Announce Type: new Abstract: Text-driven 3D scene editing with 3D Gaussian Splatting (3DGS) typically applies a 2D diffusion editor to views rendered from fixed training cameras, limiting both the spatial coverage of edits and the user's freedom to target speci…
arXiv:2607.20417v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) achieves high-quality novel-view synthesis by optimizing freely placed primitives in 3D and adaptively densifying them in under-reconstructed regions. However, this scene-adaptive capacity allocation is …
arXiv:2607.18466v1 Announce Type: new Abstract: Recent advances in differentiable Gaussian splatting have highlighted the potential of primitive-based approaches as alternative scene representations for interactive, high-quality, volume visualization (VolVis) of large datasets. H…
arXiv:2607.18801v1 Announce Type: new Abstract: Recent advancements in 3D Gaussian Splatting (3DGS) have enabled language-guided scene understanding. However, existing Referring 3D Gaussian Splatting (R3DGS) methods are fundamentally restricted to single-target queries. To reflec…
arXiv:2607.18067v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) achieves high-quality real-time rendering by representing a scene with a large collection of anisotropic Gaussian primitives. However, complex scenes often require millions of Gaussians, resulting in sub…
arXiv:2603.20714v3 Announce Type: replace Abstract: 3D Gaussian Splatting (3DGS) has become the method of choice for photo-realistic novel view synthesis (NVS), due to its efficiency and compelling visual quality. 3DGS represents the scene through a set of 3D Gaussians, parameter…
arXiv:2607.16624v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) achieves high-fidelity novel view synthesis in real-time; however its training efficiency and representation compactness are hindered by excessive primitive proliferation. To address this challenge, we f…
arXiv cs.CV
TIER_1English(EN)·Hyeonjoong Jang, Dongyoung Choi, Donggun Kim, Woohyun Kang, Min H. Kim·
arXiv:2607.16926v1 Announce Type: new Abstract: We propose a splat-based 3D scene reconstruction method from RGB-D input that effectively handles extreme motion blur, a frequent challenge in low-light environments. Under dim illumination, RGB frames often suffer from severe motio…
arXiv cs.CV
TIER_1English(EN)·Chengbo Wang, Guozheng Ma, Jinhong Wu, Tie Ji, Yizhen Lao·
arXiv:2607.17965v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) employs Gaussian primitives for explicit scene representation, facilitating real-time, high-fidelity reconstruction and novel view synthesis of complex scenes. However, the explicit modeling inherent in …
arXiv:2607.17803v1 Announce Type: new Abstract: Projector-camera (ProCams) systems achieve active scene perception and controllable appearance manipulation via structured illumination, serving as a core infrastructure for spatial augmented reality, projection mapping, and surface…
arXiv:2607.16838v1 Announce Type: new Abstract: Extracting structured, parametric 3D representations from raw images remains a fundamental challenge in computer vision and graphics. While recent advancements in the 3D Gaussian Splatting (3DGS) pipeline integrate planar primitives…
arXiv:2607.15542v1 Announce Type: new Abstract: On-the-fly reconstruction is a key requirement for many applications in robotics and autonomous navigation. Variational Bayes Gaussian Splatting (VBGS) enables continual learning without replay buffers using Coordinate Ascent Variat…
arXiv:2508.18242v2 Announce Type: replace Abstract: We introduce GSVisLoc, a visual localization method designed for 3D Gaussian Splatting (3DGS) scene representations. Given a 3DGS model of a scene and a query image, our goal is to estimate the camera's position and orientation.…
arXiv cs.CV
TIER_1English(EN)·Joanna Kaleta, Bartosz \'Swirta, Kacper Kania, Tomasz Trzci\'nski, Przemys{\l}aw Spurek, Marek Kowalski·
arXiv:2602.04043v2 Announce Type: replace Abstract: The growing demand for rapid and scalable 3D asset creation has driven interest in feed-forward 3D reconstruction methods, with 3D Gaussian Splatting (3DGS) emerging as an effective scene representation. While recent approaches …
arXiv:2607.14990v1 Announce Type: new Abstract: When a camera moves fast during exposure, blur destroys the intra-exposure motion a 3D model needs to recover the sharp scene, while event cameras capture exactly this signal at microsecond resolution. Turning them into reliable 3D …
arXiv:2607.14513v1 Announce Type: new Abstract: Techniques for modeling 3D scenes from image collections, such as 3D Gaussian Splatting (3DGS), are capable of generating high-quality novel views by leveraging graphics primitives with view-dependent appearance. In 3DGS, spherical …
arXiv cs.CV
TIER_1English(EN)·Andreas Meuleman, Linus Franke, Boris Zhestiankin, Camille Montemagni, George Drettakis·
arXiv:2607.14481v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) has become the method of choice for reconstructing and real-time rendering of captured scenes. To capture a scene with good visual quality, continuous image sequences are usually combined with out-of-ord…
When a camera moves fast during exposure, blur destroys the intra-exposure motion a 3D model needs to recover the sharp scene, while event cameras capture exactly this signal at microsecond resolution. Turning them into reliable 3D supervision faces two obstacles. First, the two …
arXiv cs.CV
TIER_1English(EN)·Chulin Zhao, Yiran Xu, Shu Liu·
arXiv:2607.13682v1 Announce Type: new Abstract: Radiative Gaussian splatting has made sparse-view CT reconstruction fast, but existing methods output point estimates with no notion of where the reconstruction can be trusted. We exploit a property of transmissive X-ray imaging tha…
arXiv cs.CV
TIER_1English(EN)·Butian Xiong, Rong Liu, Tiantian Zhou, Meida Chen, Zhiwen Fan, Andrew Feng·
arXiv:2603.16103v2 Announce Type: replace Abstract: 3D Gaussian Splat (3DGS) enables high-fidelity, real-time novel view synthesis by representing scenes with large sets of anisotropic primitives, but often requires millions of Splats, incurring significant storage and transmissi…
Radiative Gaussian splatting has made sparse-view CT reconstruction fast, but existing methods output point estimates with no notion of where the reconstruction can be trusted. We exploit a property of transmissive X-ray imaging that RGB splatting cannot claim -- projection and v…
arXiv cs.CV
TIER_1English(EN)·Seung-gyeom Kim, Areum Kim, Yongjae Yoo, Sukmin Yun·
arXiv:2607.12362v1 Announce Type: new Abstract: Recent 4D Gaussian Splatting (4DGS) methods often fail under fast motion with large inter-frame displacements, where Gaussian attributes are poorly learned during training, and fast-moving objects are often lost from the reconstruct…
arXiv:2607.12785v1 Announce Type: new Abstract: Robot-assisted minimally invasive surgery (MIS) critically depends on reliable endoscopic perception for navigation and safety. However, conventional endoscopes provide only a limited field of view, leaving large portions of surroun…
Robot-assisted minimally invasive surgery (MIS) critically depends on reliable endoscopic perception for navigation and safety. However, conventional endoscopes provide only a limited field of view, leaving large portions of surrounding anatomy unobserved. Recent neural rendering…
Recent 4D Gaussian Splatting (4DGS) methods often fail under fast motion with large inter-frame displacements, where Gaussian attributes are poorly learned during training, and fast-moving objects are often lost from the reconstruction. In this work, we introduce Spatiotemporal P…
arXiv:2603.14316v2 Announce Type: replace Abstract: Object-level 3D reconstruction play important roles across domains such as cultural heritage digitization, industrial manufacturing, and virtual reality. However, existing Gaussian Splatting-based approaches generally rely on fu…
arXiv:2412.02245v3 Announce Type: replace Abstract: Recently, several studies have combined Gaussian Splatting to obtain scene representations with language embeddings for open-vocabulary 3D scene understanding. While these methods perform well, they essentially require very dens…
arXiv cs.CV
TIER_1English(EN)·Yingji Zhong, Dave Zhenyu Chen, Fuzhao Ou, Youyu Chen, Zhihao Li, Lanqing Hong, Dan Xu·
arXiv:2607.10995v1 Announce Type: new Abstract: Recent generalizable 3D Gaussian Splatting models have advanced long-sequence novel view synthesis (NVS), but at the cost of substantial redundant computation. We identify that the redundancy can be mitigated based on two observatio…
🔬 New W29 tutorial: 3D Gaussian Splatting gives you one best-fit scene and no idea where it's trustworthy. Turn point-estimate splats into distributions over geometry: • Bayes' Rays — post-hoc Laplace posterior • FisherRF — Fisher info picks the next view • diffusion/score priors…
