You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
<img data-src="assets/thumbnails/xu2024gaussianproperty.jpg" data-fallback="None" alt="Paper thumbnail for GaussianProperty: Integrating Physical Properties to 3D Gaussians with LMMs" class="lazy" loading="lazy"/>
2300
+
</div>
2301
+
<div class="paper-content">
2302
+
<h2 class="paper-title">GaussianProperty: Integrating Physical Properties to 3D Gaussians with LMMs <span class="paper-year">(2024)</span></h2>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
2312
+
<div class="paper-abstract">Estimating physical properties for visual data is a crucial task in computer vision, graphics, and robotics, underpinning applications such as augmented reality, physical simulation, and robotic grasping. However, this area remains under-explored due to the inherent ambiguities in physical property estimation. To address these challenges, we introduce GaussianProperty, a training-free framework that assigns physical properties of materials to 3D Gaussians. Specifically, we integrate the segmentation capability of SAM with the recognition capability of GPT-4V(ision) to formulate a global-local physical property reasoning module for 2D images. Then we project the physical properties from multi-view 2D images to 3D Gaussians using a voting strategy. We demonstrate that 3D Gaussians with physical property annotations enable applications in physics-based dynamic simulation and robotic grasping. For physics-based dynamic simulation, we leverage the Material Point Method (MPM) for realistic dynamic simulation. For robot grasping, we develop a grasping force prediction strategy that estimates a safe force range required for object grasping based on the estimated physical properties. Extensive experiments on material segmentation, physics-based dynamic simulation, and robotic grasping validate the effectiveness of our proposed method, highlighting its crucial role in understanding physical properties from visual data. Online demo, code, more cases and annotated datasets are available on \href{https://Gaussian-Property.github.io}{this https URL}.
2313
+
</div></div>
2314
+
</div>
2315
+
</div>
2316
+
</div>
2294
2317
<div class="paper-row" data-id="liang2024supergseg" data-title="SuperGSeg: Open-Vocabulary 3D Segmentation with Structured Super-Gaussians" data-authors="Siyun Liang, Sen Wang, Kunyi Li, Michael Niemeyer, Stefano Gasperini, Nassir Navab, Federico Tombari" data-year="2024" data-tags='["Language Embedding", "Project", "Segmentation"]'>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
2721
+
<div class="paper-abstract">Recent advancements in static feed-forward scene reconstruction have demonstrated significant progress in high-quality novel view synthesis. However, these models often struggle with generalizability across diverse environments and fail to effectively handle dynamic content. We present BTimer (short for BulletTimer), the first motion-aware feed-forward model for real-time reconstruction and novel view synthesis of dynamic scenes. Our approach reconstructs the full scene in a 3D Gaussian Splatting representation at a given target ('bullet') timestamp by aggregating information from all the context frames. Such a formulation allows BTimer to gain scalability and generalization by leveraging both static and dynamic scene datasets. Given a casual monocular dynamic video, BTimer reconstructs a bullet-time scene within 150ms while reaching state-of-the-art performance on both static and dynamic scene datasets, even compared with optimization-based approaches.
2722
+
</div></div>
2723
+
</div>
2724
+
</div>
2725
+
</div>
2726
+
<div class="paper-row" data-id="tan2024planarsplatting" data-title="PlanarSplatting: Accurate Planar Surface Reconstruction in 3 Minutes" data-authors="Bin Tan, Rui Yu, Yujun Shen, Nan Xue" data-year="2024" data-tags='["Acceleration", "Project", "Rendering"]'>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
2742
+
<div class="paper-abstract">This paper presents PlanarSplatting, an ultra-fast and accurate surface reconstruction approach for multiview indoor images. We take the 3D planes as the main objective due to their compactness and structural expressiveness in indoor scenes, and develop an explicit optimization framework that learns to fit the expected surface of indoor scenes by splatting the 3D planes into 2.5D depth and normal maps. As our PlanarSplatting operates directly on the 3D plane primitives, it eliminates the dependencies on 2D/3D plane detection and plane matching and tracking for planar surface reconstruction. Furthermore, the essential merits of plane-based representation plus CUDA-based implementation of planar splatting functions, PlanarSplatting reconstructs an indoor scene in 3 minutes while having significantly better geometric accuracy. Thanks to our ultra-fast reconstruction speed, the largest quantitative evaluation on the ScanNet and ScanNet++ datasets over hundreds of scenes clearly demonstrated the advantages of our method. We believe that our accurate and ultrafast planar surface reconstruction method will be applied in the structured data curation for surface reconstruction in the future. The code of our CUDA implementation will be publicly available. Project page: https://icetttb.github.io/PlanarSplatting/
2743
+
</div></div>
2744
+
</div>
2745
+
</div>
2746
+
</div>
2747
+
<div class="paper-row" data-id="schmidt2024nerf" data-title="NeRF and Gaussian Splatting SLAM in the Wild" data-authors="Fabian Schmidt, Markus Enzweiler, Abhinav Valada" data-year="2024" data-tags='["Code", "In the Wild", "Review", "SLAM"]'>
<img data-src="assets/thumbnails/schmidt2024nerf.jpg" data-fallback="None" alt="Paper thumbnail for NeRF and Gaussian Splatting SLAM in the Wild" class="lazy" loading="lazy"/>
2753
+
</div>
2754
+
<div class="paper-content">
2755
+
<h2 class="paper-title">NeRF and Gaussian Splatting SLAM in the Wild <span class="paper-year">(2024)</span></h2>
2756
+
<p class="paper-authors">Fabian Schmidt, Markus Enzweiler, Abhinav Valada</p>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
2764
+
<div class="paper-abstract">Navigating outdoor environments with visual Simultaneous Localization and Mapping (SLAM) systems poses significant challenges due to dynamic scenes, lighting variations, and seasonal changes, requiring robust solutions. While traditional SLAM methods struggle with adaptability, deep learning-based approaches and emerging neural radiance fields as well as Gaussian Splatting-based SLAM methods, offer promising alternatives. However, these methods have primarily been evaluated in controlled indoor environments with stable conditions, leaving a gap in understanding their performance in unstructured and variable outdoor settings. This study addresses this gap by evaluating these methods in natural outdoor environments, focusing on camera tracking accuracy, robustness to environmental factors, and computational efficiency, highlighting distinct trade-offs. Extensive evaluations demonstrate that neural SLAM methods achieve superior robustness, particularly under challenging conditions such as low light, but at a high computational cost. At the same time, traditional methods perform the best across seasons but are highly sensitive to variations in lighting conditions. The code of the benchmark is publicly available at https://github.com/iis-esslingen/nerf-3dgs-benchmark.
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
2784
+
<div class="paper-abstract">Recent advancements in neural rendering, particularly 2D Gaussian Splatting (2DGS), have shown promising results for jointly reconstructing fine appearance and geometry by leveraging 2D Gaussian surfels. However, current methods face significant challenges when rendering at arbitrary viewpoints, such as anti-aliasing for down-sampled rendering, and texture detail preservation for high-resolution rendering. We proposed a novel method to align the 2D surfels with texture maps and augment it with per-ray depth sorting and fisher-based pruning for rendering consistency and efficiency. With correct order, per-surfel texture maps significantly improve the capabilities to capture fine details. Additionally, to render high-fidelity details in varying viewpoints, we designed a frustum-based sampling method to mitigate the aliasing artifacts. Experimental results on benchmarks and our custom texture-rich dataset demonstrate that our method surpasses existing techniques, particularly in detail preservation and anti-aliasing.
2785
+
</div></div>
2786
+
</div>
2787
+
</div>
2788
+
</div>
2683
2789
<div class="paper-row" data-id="joanna2024occams" data-title="Occam's LGS: A Simple Approach for Language Gaussian Splatting" data-authors="Jiahuan (Joanna) Cheng, Jan-Nico Zaech, Luc Van Gool, Danda Pani Paudel" data-year="2024" data-tags='["Acceleration", "Language Embedding", "Project", "Segmentation"]'>
@@ -2721,6 +2827,25 @@ <h2 class="paper-title">DynSUP: Dynamic Gaussian Splatting from An Unposed Image
2721
2827
</div>
2722
2828
</div>
2723
2829
</div>
2830
+
<div class="paper-row" data-id="hanson2024speedysplat" data-title="Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives" data-authors="Alex Hanson, Allen Tu, Geng Lin, Vasu Singla, Matthias Zwicker, Tom Goldstein" data-year="2024" data-tags='["Acceleration", "Sparse"]'>
<img data-src="assets/thumbnails/hanson2024speedysplat.jpg" data-fallback="None" alt="Paper thumbnail for Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives" class="lazy" loading="lazy"/>
2836
+
</div>
2837
+
<div class="paper-content">
2838
+
<h2 class="paper-title">Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives <span class="paper-year">(2024)</span></h2>
2839
+
<p class="paper-authors">Alex Hanson, Allen Tu, Geng Lin, Vasu Singla, Matthias Zwicker, Tom Goldstein</p>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
2844
+
<div class="paper-abstract">3D Gaussian Splatting (3D-GS) is a recent 3D scene reconstruction technique that enables real-time rendering of novel views by modeling scenes as parametric point clouds of differentiable 3D Gaussians. However, its rendering speed and model size still present bottlenecks, especially in resource-constrained settings. In this paper, we identify and address two key inefficiencies in 3D-GS, achieving substantial improvements in rendering speed, model size, and training time. First, we optimize the rendering pipeline to precisely localize Gaussians in the scene, boosting rendering speed without altering visual fidelity. Second, we introduce a novel pruning technique and integrate it into the training pipeline, significantly reducing model size and training time while further raising rendering speed. Our Speedy-Splat approach combines these techniques to accelerate average rendering speed by a drastic $6.71\times$ across scenes from the Mip-NeRF 360, Tanks & Temples, and Deep Blending datasets with $10.6\times$ fewer primitives than 3D-GS.
2845
+
</div></div>
2846
+
</div>
2847
+
</div>
2848
+
</div>
2724
2849
<div class="paper-row" data-id="pryadilshchikov2024t3dgs" data-title="T-3DGS: Removing Transient Objects for 3D Scene Reconstruction" data-authors="Vadim Pryadilshchikov, Alexander Markin, Artem Komarichev, Ruslan Rakhimov, Peter Wonka, Evgeny Burnaev" data-year="2024" data-tags='["Code", "Project", "Rendering"]'>
<img data-src="assets/thumbnails/chao2024textured.jpg" data-fallback="None" alt="Paper thumbnail for Textured Gaussians for Enhanced 3D Scene Appearance Modeling" class="lazy" loading="lazy"/>
2984
+
</div>
2985
+
<div class="paper-content">
2986
+
<h2 class="paper-title">Textured Gaussians for Enhanced 3D Scene Appearance Modeling <span class="paper-year">(2024)</span></h2>
2987
+
<p class="paper-authors">Brian Chao, Hung-Yu Tseng, Lorenzo Porzi, Chen Gao, Tuotuo Li, Qinbo Li, Ayush Saraf, Jia-Bin Huang, Johannes Kopf, Gordon Wetzstein, Changil Kim</p>
2988
+
<div class="paper-tags"><span class="paper-tag">In the Wild</span>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
2995
+
<div class="paper-abstract">3D Gaussian Splatting (3DGS) has recently emerged as a state-of-the-art 3D reconstruction and rendering technique due to its high-quality results and fast training and rendering time. However, pixels covered by the same Gaussian are always shaded in the same color up to a Gaussian falloff scaling factor. Furthermore, the finest geometric detail any individual Gaussian can represent is a simple ellipsoid. These properties of 3DGS greatly limit the expressivity of individual Gaussian primitives. To address these issues, we draw inspiration from texture and alpha mapping in traditional graphics and integrate it with 3DGS. Specifically, we propose a new generalized Gaussian appearance representation that augments each Gaussian with alpha~(A), RGB, or RGBA texture maps to model spatially varying color and opacity across the extent of each Gaussian. As such, each Gaussian can represent a richer set of texture patterns and geometric structures, instead of just a single color and ellipsoid as in naive Gaussian Splatting. Surprisingly, we found that the expressivity of Gaussians can be greatly improved by using alpha-only texture maps, and further augmenting Gaussians with RGB texture maps achieves the highest expressivity. We validate our method on a wide variety of standard benchmark datasets and our own custom captures at both the object and scene levels. We demonstrate image quality improvements over existing methods while using a similar or lower number of Gaussians.
2996
+
</div></div>
2997
+
</div>
2998
+
</div>
2999
+
</div>
3000
+
<div class="paper-row" data-id="wu2024cat4d" data-title="CAT4D: Create Anything in 4D with Multi-View Video Diffusion Models" data-authors="Rundi Wu, Ruiqi Gao, Ben Poole, Alex Trevithick, Changxi Zheng, Jonathan T. Barron, Aleksander Holynski" data-year="2024" data-tags='["Diffusion", "Dynamic", "Project"]'>
<img data-src="assets/thumbnails/wu2024cat4d.jpg" data-fallback="None" alt="Paper thumbnail for CAT4D: Create Anything in 4D with Multi-View Video Diffusion Models" class="lazy" loading="lazy"/>
3006
+
</div>
3007
+
<div class="paper-content">
3008
+
<h2 class="paper-title">CAT4D: Create Anything in 4D with Multi-View Video Diffusion Models <span class="paper-year">(2024)</span></h2>
3009
+
<p class="paper-authors">Rundi Wu, Ruiqi Gao, Ben Poole, Alex Trevithick, Changxi Zheng, Jonathan T. Barron, Aleksander Holynski</p>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
3016
+
<div class="paper-abstract">We present CAT4D, a method for creating 4D (dynamic 3D) scenes from monocular video. CAT4D leverages a multi-view video diffusion model trained on a diverse combination of datasets to enable novel view synthesis at any specified camera poses and timestamps. Combined with a novel sampling approach, this model can transform a single monocular video into a multi-view video, enabling robust 4D reconstruction via optimization of a deformable 3D Gaussian representation. We demonstrate competitive performance on novel view synthesis and dynamic scene reconstruction benchmarks, and highlight the creative capabilities for 4D scene generation from real or generated videos. See our project page for results and interactive demos: https://cat-4d.github.io/.
3017
+
</div></div>
3018
+
</div>
3019
+
</div>
3020
+
</div>
3021
+
<div class="paper-row" data-id="kang2024selfsplat" data-title="SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting" data-authors="Gyeongjin Kang, Jisang Yoo, Jihyeon Park, Seungtae Nam, Hyeonsoo Im, Sangheon Shin, Sangpil Kim, Eunbyung Park" data-year="2024" data-tags='["Code", "Feed-Forward", "Poses", "Project"]'>
<img data-src="assets/thumbnails/kang2024selfsplat.jpg" data-fallback="None" alt="Paper thumbnail for SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting" class="lazy" loading="lazy"/>
3027
+
</div>
3028
+
<div class="paper-content">
3029
+
<h2 class="paper-title">SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting <span class="paper-year">(2024)</span></h2>
3030
+
<p class="paper-authors">Gyeongjin Kang, Jisang Yoo, Jihyeon Park, Seungtae Nam, Hyeonsoo Im, Sangheon Shin, Sangpil Kim, Eunbyung Park</p>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
3039
+
<div class="paper-abstract">We propose SelfSplat, a novel 3D Gaussian Splatting model designed to perform pose-free and 3D prior-free generalizable 3D reconstruction from unposed multi-view images. These settings are inherently ill-posed due to the lack of ground-truth data, learned geometric information, and the need to achieve accurate 3D reconstruction without finetuning, making it difficult for conventional methods to achieve high-quality results. Our model addresses these challenges by effectively integrating explicit 3D representations with self-supervised depth and pose estimation techniques, resulting in reciprocal improvements in both pose accuracy and 3D reconstruction quality. Furthermore, we incorporate a matching-aware pose estimation network and a depth refinement module to enhance geometry consistency across views, ensuring more accurate and stable 3D reconstructions. To present the performance of our method, we evaluated it on large-scale real-world datasets, including RealEstate10K, ACID, and DL3DV. SelfSplat achieves superior results over previous state-of-the-art methods in both appearance and geometry quality, also demonstrates strong cross-dataset generalization capabilities. Extensive ablation studies and analysis also validate the effectiveness of our proposed methods. Code and pretrained models are available at https://gynjn.github.io/selfsplat/
3040
+
</div></div>
3041
+
</div>
3042
+
</div>
3043
+
</div>
3044
+
<div class="paper-row" data-id="flynn2024quark" data-title="Quark: Real-time, High-resolution, and General Neural View Synthesis" data-authors="John Flynn, Michael Broxton, Lukas Murmann, Lucy Chai, Matthew DuVall, Clément Godard, Kathryn Heal, Srinivas Kaza, Stephen Lombardi, Xuan Luo, Supreeth Achar, Kira Prabhu, Tiancheng Sun, Lynn Tsai, Ryan Overbeck" data-year="2024" data-tags='["Feed-Forward", "Project", "Rendering", "Video"]'>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
3062
+
<div class="paper-abstract">We present a novel neural algorithm for performing high-quality, high-resolution, real-time novel view synthesis. From a sparse set of input RGB images or videos streams, our network both reconstructs the 3D scene and renders novel views at 1080p resolution at 30fps on an NVIDIA A100. Our feed-forward network generalizes across a wide variety of datasets and scenes and produces state-of-the-art quality for a real-time method. Our quality approaches, and in some cases surpasses, the quality of some of the top offline methods. In order to achieve these results we use a novel combination of several key concepts, and tie them together into a cohesive and effective algorithm. We build on previous works that represent the scene using semi-transparent layers and use an iterative learned render-and-refine approach to improve those layers. Instead of flat layers, our method reconstructs layered depth maps (LDMs) that efficiently represent scenes with complex depth and occlusions. The iterative update steps are embedded in a multi-scale, UNet-style architecture to perform as much compute as possible at reduced resolution. Within each update step, to better aggregate the information from multiple input views, we use a specialized Transformer-based network component. This allows the majority of the per-input image processing to be performed in the input image space, as opposed to layer space, further increasing efficiency. Finally, due to the real-time nature of our reconstruction and rendering, we dynamically create and discard the internal 3D geometry for each frame, generating the LDM for each view. Taken together, this produces a novel and effective algorithm for view synthesis. Through extensive evaluation, we demonstrate that we achieve state-of-the-art quality at real-time rates. Project page: https://quark-3d.github.io/
3063
+
</div></div>
3064
+
</div>
3065
+
</div>
3066
+
</div>
2853
3067
<div class="paper-row" data-id="hess2024splatad" data-title="SplatAD: Real-Time Lidar and Camera Rendering with 3D Gaussian Splatting for Autonomous Driving" data-authors="Georg Hess, Carl Lindström, Maryam Fatemi, Christoffer Petersson, Lennart Svensson" data-year="2024" data-tags='["Autonomous Driving", "Project"]'>
<img data-src="assets/thumbnails/chou2024generating.jpg" data-fallback="None" alt="Paper thumbnail for Generating 3D-Consistent Videos from Unposed Internet Photos" class="lazy" loading="lazy"/>
3161
+
</div>
3162
+
<div class="paper-content">
3163
+
<h2 class="paper-title">Generating 3D-Consistent Videos from Unposed Internet Photos <span class="paper-year">(2024)</span></h2>
3164
+
<p class="paper-authors">Gene Chou, Kai Zhang, Sai Bi, Hao Tan, Zexiang Xu, Fujun Luan, Bharath Hariharan, Noah Snavely</p>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
3173
+
<div class="paper-abstract">We address the problem of generating videos from unposed internet photos. A handful of input images serve as keyframes, and our model interpolates between them to simulate a path moving between the cameras. Given random images, a model's ability to capture underlying geometry, recognize scene identity, and relate frames in terms of camera position and orientation reflects a fundamental understanding of 3D structure and scene layout. However, existing video models such as Luma Dream Machine fail at this task. We design a self-supervised method that takes advantage of the consistency of videos and variability of multiview internet photos to train a scalable, 3D-aware video model without any 3D annotations such as camera parameters. We validate that our method outperforms all baselines in terms of geometric and appearance consistency. We also show our model benefits applications that enable camera control, such as 3D Gaussian Splatting. Our results suggest that we can scale up scene-level 3D learning using only 2D data such as videos and multiview internet photos.
3174
+
</div></div>
3175
+
</div>
3176
+
</div>
3177
+
</div>
3178
+
<div class="paper-row" data-id="fang2024minisplatting2" data-title="Mini-Splatting2: Building 360 Scenes within Minutes via Aggressive Gaussian Densification" data-authors="Guangchi Fang, Bing Wang" data-year="2024" data-tags='["Acceleration", "Densification"]'>
<img data-src="assets/thumbnails/fang2024minisplatting2.jpg" data-fallback="None" alt="Paper thumbnail for Mini-Splatting2: Building 360 Scenes within Minutes via Aggressive Gaussian Densification" class="lazy" loading="lazy"/>
3184
+
</div>
3185
+
<div class="paper-content">
3186
+
<h2 class="paper-title">Mini-Splatting2: Building 360 Scenes within Minutes via Aggressive Gaussian Densification <span class="paper-year">(2024)</span></h2>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
3192
+
<div class="paper-abstract">In this study, we explore the essential challenge of fast scene optimization for Gaussian Splatting. Through a thorough analysis of the geometry modeling process, we reveal that dense point clouds can be effectively reconstructed early in optimization through Gaussian representations. This insight leads to our approach of aggressive Gaussian densification, which provides a more efficient alternative to conventional progressive densification methods. By significantly increasing the number of critical Gaussians, we enhance the model capacity to capture dense scene geometry at the early stage of optimization. This strategy is seamlessly integrated into the Mini-Splatting densification and simplification framework, enabling rapid convergence without compromising quality. Additionally, we introduce visibility culling within Gaussian Splatting, leveraging per-view Gaussian importance as precomputed visibility to accelerate the optimization process. Our Mini-Splatting2 achieves a balanced trade-off among optimization time, the number of Gaussians, and rendering quality, establishing a strong baseline for future Gaussian-Splatting-based works. Our work sets the stage for more efficient, high-quality 3D scene modeling in real-world applications, and the code will be made available no matter acceptance.
3193
+
</div></div>
3194
+
</div>
3195
+
</div>
3196
+
</div>
3197
+
<div class="paper-row" data-id="zhou2024gpsgaussian" data-title="GPS-Gaussian+: Generalizable Pixel-wise 3D Gaussian Splatting for Real-Time Human-Scene Rendering from Sparse Views" data-authors="Boyao Zhou, Shunyuan Zheng, Hanzhang Tu, Ruizhi Shao, Boning Liu, Shengping Zhang, Liqiang Nie, Yebin Liu" data-year="2024" data-tags='["Acceleration", "Dynamic", "Project", "Rendering"]'>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
3214
+
<div class="paper-abstract">Differentiable rendering techniques have recently shown promising results for free-viewpoint video synthesis of characters. However, such methods, either Gaussian Splatting or neural implicit rendering, typically necessitate per-subject optimization which does not meet the requirement of real-time rendering in an interactive application. We propose a generalizable Gaussian Splatting approach for high-resolution image rendering under a sparse-view camera setting. To this end, we introduce Gaussian parameter maps defined on the source views and directly regress Gaussian properties for instant novel view synthesis without any fine-tuning or optimization. We train our Gaussian parameter regression module on human-only data or human-scene data, jointly with a depth estimation module to lift 2D parameter maps to 3D space. The proposed framework is fully differentiable with both depth and rendering supervision or with only rendering supervision. We further introduce a regularization term and an epipolar attention mechanism to preserve geometry consistency between two source views, especially when neglecting depth supervision. Experiments on several datasets demonstrate that our method outperforms state-of-the-art methods while achieving an exceeding rendering speed.
<img data-src="assets/thumbnails/tang2024spars3r.jpg" data-fallback="None" alt="Paper thumbnail for SPARS3R: Semantic Prior Alignment and Regularization for Sparse 3D Reconstruction" class="lazy" loading="lazy"/>
3244
+
</div>
3245
+
<div class="paper-content">
3246
+
<h2 class="paper-title">SPARS3R: Semantic Prior Alignment and Regularization for Sparse 3D Reconstruction <span class="paper-year">(2024)</span></h2>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
3253
+
<div class="paper-abstract">Recent efforts in Gaussian-Splat-based Novel View Synthesis can achieve photorealistic rendering; however, such capability is limited in sparse-view scenarios due to sparse initialization and over-fitting floaters. Recent progress in depth estimation and alignment can provide dense point cloud with few views; however, the resulting pose accuracy is suboptimal. In this work, we present SPARS3R, which combines the advantages of accurate pose estimation from Structure-from-Motion and dense point cloud from depth estimation. To this end, SPARS3R first performs a Global Fusion Alignment process that maps a prior dense point cloud to a sparse point cloud from Structure-from-Motion based on triangulated correspondences. RANSAC is applied during this process to distinguish inliers and outliers. SPARS3R then performs a second, Semantic Outlier Alignment step, which extracts semantically coherent regions around the outliers and performs local alignment in these regions. Along with several improvements in the evaluation process, we demonstrate that SPARS3R can achieve photorealistic rendering with sparse images and significantly outperforms existing approaches.
3254
+
</div></div>
3255
+
</div>
3256
+
</div>
3257
+
</div>
2960
3258
<div class="paper-row" data-id="svitov2024billboard" data-title="BillBoard Splatting (BBSplat): Learnable Textured Primitives for Novel View Synthesis" data-authors="David Svitov, Pietro Morerio, Lourdes Agapito, Alessio Del Bue" data-year="2024" data-tags='["Code", "Optimization", "Project", "Texturing", "Video"]'>
<img data-src="assets/thumbnails/zhang2024gaussianspa.jpg" data-fallback="None" alt="Paper thumbnail for GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting" class="lazy" loading="lazy"/>
3308
+
</div>
3309
+
<div class="paper-content">
3310
+
<h2 class="paper-title">GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting <span class="paper-year">(2024)</span></h2>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
3318
+
<div class="paper-abstract">3D Gaussian Splatting (3DGS) has emerged as a mainstream for novel view synthesis, leveraging continuous aggregations of Gaussian functions to model scene geometry. However, 3DGS suffers from substantial memory requirements to store the multitude of Gaussians, hindering its practicality. To address this challenge, we introduce GaussianSpa, an optimization-based simplification framework for compact and high-quality 3DGS. Specifically, we formulate the simplification as an optimization problem associated with the 3DGS training. Correspondingly, we propose an efficient "optimizing-sparsifying" solution that alternately solves two independent sub-problems, gradually imposing strong sparsity onto the Gaussians in the training process. Our comprehensive evaluations on various datasets show the superiority of GaussianSpa over existing state-of-the-art approaches. Notably, GaussianSpa achieves an average PSNR improvement of 0.9 dB on the real-world Deep Blending dataset with 10$\times$ fewer Gaussians compared to the vanilla 3DGS. Our project page is available at https://gaussianspa.github.io/.
3319
+
</div></div>
3320
+
</div>
3321
+
</div>
3322
+
</div>
3004
3323
<div class="paper-row" data-id="lu20243dgscd" data-title="3DGS-CD: 3D Gaussian Splatting-based Change Detection for Physical Object Rearrangement" data-authors="Ziqi Lu, Jianbo Ye, John Leonard" data-year="2024" data-tags='["Robotics"]'>
<img data-src="assets/thumbnails/hou2024sortfree.jpg" data-fallback="None" alt="Paper thumbnail for Sort-free Gaussian Splatting via Weighted Sum Rendering" class="lazy" loading="lazy"/>
3434
+
</div>
3435
+
<div class="paper-content">
3436
+
<h2 class="paper-title">Sort-free Gaussian Splatting via Weighted Sum Rendering <span class="paper-year">(2024)</span></h2>
3437
+
<p class="paper-authors">Qiqi Hou, Randall Rauwendaal, Zifeng Li, Hoang Le, Farzad Farhadzadeh, Fatih Porikli, Alexei Bourd, Amir Said</p>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
3442
+
<div class="paper-abstract">Recently, 3D Gaussian Splatting (3DGS) has emerged as a significant advancement in 3D scene reconstruction, attracting considerable attention due to its ability to recover high-fidelity details while maintaining low complexity. Despite the promising results achieved by 3DGS, its rendering performance is constrained by its dependence on costly non-commutative alpha-blending operations. These operations mandate complex view dependent sorting operations that introduce computational overhead, especially on the resource-constrained platforms such as mobile phones. In this paper, we propose Weighted Sum Rendering, which approximates alpha blending with weighted sums, thereby removing the need for sorting. This simplifies implementation, delivers superior performance, and eliminates the "popping" artifacts caused by sorting. Experimental results show that optimizing a generalized Gaussian splatting formulation to the new differentiable rendering yields competitive image quality. The method was implemented and tested in a mobile device GPU, achieving on average $1.23\times$ faster rendering.
<img data-src="assets/thumbnails/xie2024supergs.jpg" data-fallback="None" alt="Paper thumbnail for SuperGS: Super-Resolution 3D Gaussian Splatting via Latent Feature Field and Gradient-guided Splitting" class="lazy" loading="lazy"/>
3609
+
</div>
3610
+
<div class="paper-content">
3611
+
<h2 class="paper-title">SuperGS: Super-Resolution 3D Gaussian Splatting via Latent Feature Field and Gradient-guided Splitting <span class="paper-year">(2024)</span></h2>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
3618
+
<div class="paper-abstract">Recently, 3D Gaussian Splatting (3DGS) has exceled in novel view synthesis with its real-time rendering capabilities and superior quality. However, it faces challenges for high-resolution novel view synthesis (HRNVS) due to the coarse nature of primitives derived from low-resolution input views. To address this issue, we propose Super-Resolution 3DGS (SuperGS), which is an expansion of 3DGS designed with a two-stage coarse-to-fine training framework, utilizing pretrained low-resolution scene representation as an initialization for super-resolution optimization. Moreover, we introduce Multi-resolution Feature Gaussian Splatting (MFGS) to incorporates a latent feature field for flexible feature sampling and Gradient-guided Selective Splitting (GSS) for effective Gaussian upsampling. By integrating these strategies within the coarse-to-fine framework ensure both high fidelity and memory efficiency. Extensive experiments demonstrate that SuperGS surpasses state-of-the-art HRNVS methods on challenging real-world datasets using only low-resolution inputs.
3619
+
</div></div>
3620
+
</div>
3621
+
</div>
3622
+
</div>
3265
3623
<div class="paper-row" data-id="cao20243dgsdet" data-title="3DGS-DET: Empower 3D Gaussian Splatting with Boundary Guidance and Box-Focused Sampling for 3D Object Detection" data-authors="Yang Cao, Yuanliang Jv, Dan Xu" data-year="2024" data-tags='["Object Detection"]'>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
3868
+
<div class="paper-abstract">Reconstructing scenes and tracking motion are two sides of the same coin. Tracking points allow for geometric reconstruction [14], while geometric reconstruction of (dynamic) scenes allows for 3D tracking of points over time [24, 39]. The latter was recently also exploited for 2D point tracking to overcome occlusion ambiguities by lifting tracking directly into 3D [38]. However, above approaches either require offline processing or multi-view camera setups both unrealistic for real-world applications like robot navigation or mixed reality. We target the challenge of online 2D and 3D point tracking from unposed monocular camera input introducing Dynamic Online Monocular Reconstruction (DynOMo). We leverage 3D Gaussian splatting to reconstruct dynamic scenes in an online fashion. Our approach extends 3D Gaussians to capture new content and object motions while estimating camera movements from a single RGB frame. DynOMo stands out by enabling emergence of point trajectories through robust image feature reconstruction and a novel similarity-enhanced regularization term, without requiring any correspondence-level supervision. It sets the first baseline for online point tracking with monocular unposed cameras, achieving performance on par with existing methods. We aim to inspire the community to advance online point tracking and reconstruction, expanding the applicability to diverse real-world scenarios.
3869
+
</div></div>
3870
+
</div>
3871
+
</div>
3872
+
</div>
3496
3873
<div class="paper-row" data-id="chen2024omnire" data-title="OmniRe: Omni Urban Scene Reconstruction" data-authors="Ziyu Chen, Jiawei Yang, Jiahui Huang, Riccardo de Lutio, Janick Martinez Esturo, Boris Ivanovic, Or Litany, Zan Gojcic, Sanja Fidler, Marco Pavone, Li Song, Yue Wang" data-year="2024" data-tags='["Autonomous Driving", "Code", "Project"]'>
<img data-src="assets/thumbnails/zhang202425.jpg" data-fallback="None" alt="Paper thumbnail for '25] 10. TranSplat: Generalizable 3D Gaussian Splatting from Sparse Multi-View Images with Transformers" class="lazy" loading="lazy"/>
3966
+
<img data-src="assets/thumbnails/zhang202425.jpg" data-fallback="None" alt="Paper thumbnail for TranSplat: Generalizable 3D Gaussian Splatting from Sparse Multi-View Images with Transformers" class="lazy" loading="lazy"/>
3590
3967
</div>
3591
3968
<div class="paper-content">
3592
-
<h2 class="paper-title">'25] 10. TranSplat: Generalizable 3D Gaussian Splatting from Sparse Multi-View Images with Transformers <span class="paper-year">(2024)</span></h2>
3969
+
<h2 class="paper-title">TranSplat: Generalizable 3D Gaussian Splatting from Sparse Multi-View Images with Transformers <span class="paper-year">(2024)</span></h2>
<div class="paper-row" data-id="dihlmann2024subsurface" data-title="Subsurface Scattering for 3D Gaussian Splatting" data-authors="Jan-Niklas Dihlmann, Arjun Majumdar, Andreas Engelhardt, Raphael Braun, Hendrik P. A. Lensch" data-year="2024" data-tags='["Project", "Relight", "Rendering"]'>
<img data-src="assets/thumbnails/dihlmann2024subsurface.jpg" data-fallback="None" alt="Paper thumbnail for Subsurface Scattering for 3D Gaussian Splatting" class="lazy" loading="lazy"/>
3990
+
</div>
3991
+
<div class="paper-content">
3992
+
<h2 class="paper-title">Subsurface Scattering for 3D Gaussian Splatting <span class="paper-year">(2024)</span></h2>
3993
+
<p class="paper-authors">Jan-Niklas Dihlmann, Arjun Majumdar, Andreas Engelhardt, Raphael Braun, Hendrik P. A. Lensch</p>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
4000
+
<div class="paper-abstract">3D reconstruction and relighting of objects made from scattering materials present a significant challenge due to the complex light transport beneath the surface. 3D Gaussian Splatting introduced high-quality novel view synthesis at real-time speeds. While 3D Gaussians efficiently approximate an object's surface, they fail to capture the volumetric properties of subsurface scattering. We propose a framework for optimizing an object's shape together with the radiance transfer field given multi-view OLAT (one light at a time) data. Our method decomposes the scene into an explicit surface represented as 3D Gaussians, with a spatially varying BRDF, and an implicit volumetric representation of the scattering component. A learned incident light field accounts for shadowing. We optimize all parameters jointly via ray-traced differentiable rendering. Our approach enables material editing, relighting and novel view synthesis at interactive rates. We show successful application on synthetic data and introduce a newly acquired multi-view multi-light dataset of objects in a light-stage setup. Compared to previous work we achieve comparable or better results at a fraction of optimization and rendering time while enabling detailed control over material attributes. Project page https://sss.jdihlmann.com/
4001
+
</div></div>
4002
+
</div>
4003
+
</div>
4004
+
</div>
3607
4005
<div class="paper-row" data-id="liu2024gsloc" data-title="GSLoc: Efficient Camera Pose Refinement via 3D Gaussian Splatting" data-authors="Changkun Liu, Shuai Chen, Yash Bhalgat, Siyan Hu, Ming Cheng, Zirui Wang, Victor Adrian Prisacariu, Tristan Braud" data-year="2024" data-tags='["Poses", "Project"]'>
<img data-src="assets/thumbnails/zheng2024headgap.jpg" data-fallback="None" alt="Paper thumbnail for HeadGAP: Few-shot 3D Head Avatar via Generalizable Gaussian Priors" class="lazy" loading="lazy"/>
4073
+
</div>
4074
+
<div class="paper-content">
4075
+
<h2 class="paper-title">HeadGAP: Few-shot 3D Head Avatar via Generalizable Gaussian Priors <span class="paper-year">(2024)</span></h2>
4076
+
<p class="paper-authors">Xiaozheng Zheng, Chao Wen, Zhaohu Li, Weiyi Zhang, Zhuo Su, Xu Chang, Yang Zhao, Zheng Lv, Xiaoyuan Zhang, Yongjie Zhang, Guidong Wang, Lan Xu</p>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
4083
+
<div class="paper-abstract">In this paper, we present a novel 3D head avatar creation approach capable of generalizing from few-shot in-the-wild data with high-fidelity and animatable robustness. Given the underconstrained nature of this problem, incorporating prior knowledge is essential. Therefore, we propose a framework comprising prior learning and avatar creation phases. The prior learning phase leverages 3D head priors derived from a large-scale multi-view dynamic dataset, and the avatar creation phase applies these priors for few-shot personalization. Our approach effectively captures these priors by utilizing a Gaussian Splatting-based auto-decoder network with part-based dynamic modeling. Our method employs identity-shared encoding with personalized latent codes for individual identities to learn the attributes of Gaussian primitives. During the avatar creation phase, we achieve fast head avatar personalization by leveraging inversion and fine-tuning strategies. Extensive experiments demonstrate that our model effectively exploits head priors and successfully generalizes them to few-shot personalization, achieving photo-realistic rendering quality, multi-view consistency, and stable animation.
4084
+
</div></div>
4085
+
</div>
4086
+
</div>
4087
+
</div>
3669
4088
<div class="paper-row" data-id="chahe2024query3d" data-title="Query3D: LLM-Powered Open-Vocabulary Scene Segmentation with Language Embedded 3D Gaussian" data-authors="Amirhosein Chahe, Lifeng Zhou" data-year="2024" data-tags='["Code", "Language Embedding", "Segmentation"]'>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
4167
+
<div class="paper-abstract">3D Gaussian Splatting (3DGS) has emerged as a prominent technique with the potential to become a mainstream method for 3D representations. It can effectively transform multi-view images into explicit 3D Gaussian through efficient training, and achieve real-time rendering of novel views. This survey aims to analyze existing 3DGS-related works from multiple intersecting perspectives, including related tasks, technologies, challenges, and opportunities. The primary objective is to provide newcomers with a rapid understanding of the field and to assist researchers in methodically organizing existing technologies and challenges. Specifically, we delve into the optimization, application, and extension of 3DGS, categorizing them based on their focuses or motivations. Additionally, we summarize and classify nine types of technical modules and corresponding improvements identified in existing works. Based on these analyses, we further examine the common challenges and technologies across various tasks, proposing potential research opportunities.
4168
+
</div></div>
4169
+
</div>
4170
+
</div>
4171
+
</div>
3735
4172
<div class="paper-row" data-id="moenne-loccoz20243d" data-title="3D Gaussian Ray Tracing: Fast Tracing of Particle Scenes" data-authors="Nicolas Moenne-Loccoz, Ashkan Mirzaei, Or Perel, Riccardo de Lutio, Janick Martinez Esturo, Gavriel State, Sanja Fidler, Nicholas Sharp, Zan Gojcic" data-year="2024" data-tags='["Project", "Ray Tracing", "Video"]'>
<img data-src="assets/thumbnails/li2024gsoctree.jpg" data-fallback="None" alt="Paper thumbnail for GS-Octree: Octree-based 3D Gaussian Splatting for Robust Object-level 3D Reconstruction Under Strong Lighting" class="lazy" loading="lazy"/>
4325
+
</div>
4326
+
<div class="paper-content">
4327
+
<h2 class="paper-title">GS-Octree: Octree-based 3D Gaussian Splatting for Robust Object-level 3D Reconstruction Under Strong Lighting <span class="paper-year">(2024)</span></h2>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
4332
+
<div class="paper-abstract">The 3D Gaussian Splatting technique has significantly advanced the construction of radiance fields from multi-view images, enabling real-time rendering. While point-based rasterization effectively reduces computational demands for rendering, it often struggles to accurately reconstruct the geometry of the target object, especially under strong lighting. To address this challenge, we introduce a novel approach that combines octree-based implicit surface representations with Gaussian splatting. Our method consists of four stages. Initially, it reconstructs a signed distance field (SDF) and a radiance field through volume rendering, encoding them in a low-resolution octree. The initial SDF represents the coarse geometry of the target object. Subsequently, it introduces 3D Gaussians as additional degrees of freedom, which are guided by the SDF. In the third stage, the optimized Gaussians further improve the accuracy of the SDF, allowing it to recover finer geometric details compared to the initial SDF obtained in the first stage. Finally, it adopts the refined SDF to further optimize the 3D Gaussians via splatting, eliminating those that contribute little to visual appearance. Experimental results show that our method, which leverages the distribution of 3D Gaussians with SDFs, reconstructs more accurate geometry, particularly in images with specular highlights caused by strong lighting.
4333
+
</div></div>
4334
+
</div>
4335
+
</div>
4336
+
</div>
3882
4337
<div class="paper-row" data-id="papantonakis2024reducing" data-title="Reducing the Memory Footprint of 3D Gaussian Splatting" data-authors="Panagiotis Papantonakis, Georgios Kopanas, Bernhard Kerbl, Alexandre Lanvin, George Drettakis" data-year="2024" data-tags='["Code", "Compression", "Project", "Video"]'>
<img data-src="assets/thumbnails/li2024garmentdreamer.jpg" data-fallback="None" alt="Paper thumbnail for GarmentDreamer: 3DGS Guided Garment Synthesis with Diverse Geometry and Texture Details" class="lazy" loading="lazy"/>
4841
+
</div>
4842
+
<div class="paper-content">
4843
+
<h2 class="paper-title">GarmentDreamer: 3DGS Guided Garment Synthesis with Diverse Geometry and Texture Details <span class="paper-year">(2024)</span></h2>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
4857
+
<div class="paper-abstract">Traditional 3D garment creation is labor-intensive, involving sketching, modeling, UV mapping, and texturing, which are time-consuming and costly. Recent advances in diffusion-based generative models have enabled new possibilities for 3D garment generation from text prompts, images, and videos. However, existing methods either suffer from inconsistencies among multi-view images or require additional processes to separate cloth from the underlying human model. In this paper, we propose GarmentDreamer, a novel method that leverages 3D Gaussian Splatting (GS) as guidance to generate wearable, simulation-ready 3D garment meshes from text prompts. In contrast to using multi-view images directly predicted by generative models as guidance, our 3DGS guidance ensures consistent optimization in both garment deformation and texture synthesis. Our method introduces a novel garment augmentation module, guided by normal and RGBA information, and employs implicit Neural Texture Fields (NeTF) combined with Score Distillation Sampling (SDS) to generate diverse geometric and texture details. We validate the effectiveness of our approach through comprehensive qualitative and quantitative experiments, showcasing the superior performance of GarmentDreamer over state-of-the-art alternatives. Our project page is available at: https://xuan-li.github.io/GarmentDreamerDemo/.
4858
+
</div></div>
4859
+
</div>
4860
+
</div>
4861
+
</div>
4862
+
<div class="paper-row" data-id="wu2024gaussian" data-title="Gaussian Head & Shoulders: High Fidelity Neural Upper Body Avatars with Anchor Gaussian Guided Texture Warping" data-authors="Tianhao Wu, Jing Yang, Zhilin Guo, Jingyi Wan, Fangcheng Zhong, Cengiz Oztireli" data-year="2024" data-tags='["Avatar", "Dynamic", "Project"]'>
<img data-src="assets/thumbnails/wu2024gaussian.jpg" data-fallback="None" alt="Paper thumbnail for Gaussian Head & Shoulders: High Fidelity Neural Upper Body Avatars with Anchor Gaussian Guided Texture Warping" class="lazy" loading="lazy"/>
4868
+
</div>
4869
+
<div class="paper-content">
4870
+
<h2 class="paper-title">Gaussian Head & Shoulders: High Fidelity Neural Upper Body Avatars with Anchor Gaussian Guided Texture Warping <span class="paper-year">(2024)</span></h2>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
4878
+
<div class="paper-abstract">By equipping the most recent 3D Gaussian Splatting representation with head 3D morphable models (3DMM), existing methods manage to create head avatars with high fidelity. However, most existing methods only reconstruct a head without the body, substantially limiting their application scenarios. We found that naively applying Gaussians to model the clothed chest and shoulders tends to result in blurry reconstruction and noisy floaters under novel poses. This is because of the fundamental limitation of Gaussians and point clouds -- each Gaussian or point can only have a single directional radiance without spatial variance, therefore an unnecessarily large number of them is required to represent complicated spatially varying texture, even for simple geometry. In contrast, we propose to model the body part with a neural texture that consists of coarse and pose-dependent fine colors. To properly render the body texture for each view and pose without accurate geometry nor UV mapping, we optimize another sparse set of Gaussians as anchors that constrain the neural warping field that maps image plane coordinates to the texture space. We demonstrate that Gaussian Head & Shoulders can fit the high-frequency details on the clothed upper body with high fidelity and potentially improve the accuracy and fidelity of the head region. We evaluate our method with casual phone-captured and internet videos and show our method archives superior reconstruction quality and robustness in both self and cross reenactment tasks. To fully utilize the efficient rendering speed of Gaussian splatting, we additionally propose an accelerated inference method of our trained model without Multi-Layer Perceptron (MLP) queries and reach a stable rendering speed of around 130 FPS for any subjects.
4879
+
</div></div>
4880
+
</div>
4881
+
</div>
4882
+
</div>
4380
4883
<div class="paper-row" data-id="dalal2024gaussian" data-title="Gaussian Splatting: 3D Reconstruction and Novel View Synthesis, a Review" data-authors="Anurag Dalal, Daniel Hagen, Kjell G. Robbersmyr, Kristian Muri Knausgård" data-year="2024" data-tags='["Review"]'>
<button class="abstract-toggle" onclick="toggleAbstract(this)">📖 Show Abstract</button>
4832
5337
<div class="paper-abstract">In this paper, we propose a 3D geometry-aware deformable Gaussian Splatting method for dynamic view synthesis. Existing neural radiance fields (NeRF) based solutions learn the deformation in an implicit manner, which cannot incorporate 3D scene geometry. Therefore, the learned deformation is not necessarily geometrically coherent, which results in unsatisfactory dynamic view synthesis and 3D dynamic reconstruction. Recently, 3D Gaussian Splatting provides a new representation of the 3D scene, building upon which the 3D geometry could be exploited in learning the complex 3D deformation. Specifically, the scenes are represented as a collection of 3D Gaussian, where each 3D Gaussian is optimized to move and rotate over time to model the deformation. To enforce the 3D scene geometry constraint during deformation, we explicitly extract 3D geometry features and integrate them in learning the 3D deformation. In this way, our solution achieves 3D geometry-aware deformation modeling, which enables improved dynamic view synthesis and 3D dynamic reconstruction. Extensive experimental results on both synthetic and real datasets prove the superiority of our solution, which achieves new state-of-the-art performance.</div></div>
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?v=4&size=40){kind=avatar}
0 commit comments