Compact 3D Gaussian Representation for Radiance Field
Authors: Joo Chan Lee, Daniel Rho, Xiangyu Sun, Jong Hwan Ko, Eunbyung Park
Abstract
Neural Radiance Fields (NeRFs) have demonstrated remarkable potential in capturing complex 3D scenes with high fidelity. However, one persistent challenge that hinders the widespread adoption of NeRFs is the computational bottleneck due to the volumetric rendering. On the other hand, 3D Gaussian splatting (3DGS) has recently emerged as an alternative representation that leverages a 3D Gaussisan-based representation and adopts the rasterization pipeline to render the images rather than volumetric rendering, achieving very fast rendering speed and promising image quality. However, a significant drawback arises as 3DGS entails a substantial number of 3D Gaussians to maintain the high fidelity of the rendered images, which requires a large amount of memory and storage. To address this critical issue, we place a specific emphasis on two key objectives: reducing the number of Gaussian points without sacrificing performance and compressing the Gaussian attributes, such as view-dependent color and covariance. To this end, we propose a learnable mask strategy that significantly reduces the number of Gaussians while preserving high performance. In addition, we propose a compact but effective representation of view-dependent color by employing a grid-based neural field rather than relying on spherical harmonics. Finally, we learn codebooks to compactly represent the geometric attributes of Gaussian by vector quantization. In our extensive experiments, we consistently show over 10$\times$ reduced storage and enhanced rendering speed, while maintaining the quality of the scene representation, compared to 3DGS. Our work provides a comprehensive framework for 3D scene representation, achieving high performance, fast training, compactness, and real-time rendering. Our project page is available at https://maincold2.github.io/c3dgs/.
The papers around it, by date and closeness
Higher is closer to this paper; left is older. Lines join papers close to each other: groups of lines are sub-topics. Select a point to see why it is there. Back to the summary of this paper
Papers around this one
Ranked by how close their title and abstract are to this paper’s. Select some to read them together.
Closest papers
-
A Refined 3D Gaussian Representation for High-Quality Dynamic Scene ReconstructionBin Zhang, Bi Zeng, Zexin Peng · 2024 · IEEE International Conference on Robotics and Automation · cited 1 time · Not summarized yet
A Refined 3D Gaussian Representation targets high-quality dynamic-scene reconstruction while addressing the memory costs of Gaussian representations.
Abstract
In recent years, Neural Radiance Fields (NeRF) has revolutionized three-dimensional (3D) reconstruction with its implicit representation. Building upon NeRF, 3D Gaussian Splatting (3D-GS) has departed from the implicit representation of neural networks and instead directly represents scenes as point clouds with Gaussian-shaped distributions. While this shift has notably elevated the rendering quality and speed of radiance fields but inevitably led to a significant increase in memory usage. Additionally, effectively rendering dynamic scenes in 3D-GS has emerged as a pressing challenge. To address these concerns, this paper proposes a refined 3D Gaussian representation for high-quality dynamic…
-
EAGLES: Efficient Accelerated 3D Gaussians with Lightweight EncodingSSharath Girish, Kamal Gupta, Abhinav Shrivastava · 2023 · Computer Vision and Pattern Recognition · cited 300 times · AI summary available
EAGLES compresses 3D Gaussian storage with quantized embeddings and coarse-to-fine training, addressing the same memory and optimization costs.
Abstract
Recently, 3D Gaussian splatting (3D-GS) has gained popularity in novel-view scene synthesis. It addresses the challenges of lengthy training times and slow rendering speeds associated with Neural Radiance Fields (NeRFs). Through rapid, differentiable rasterization of 3D Gaussians, 3D-GS achieves real-time rendering and accelerated training. They, however, demand substantial memory resources for both training and storage, as they require millions of Gaussians in their point cloud representation for each scene. We present a technique utilizing quantized embeddings to significantly reduce memory storage requirements and a coarse-to-fine training strategy for a faster and more stable optimizatio…
-
HyRF: Hybrid Radiance Fields for Memory-efficient and High-quality Novel View SynthesisZi-Peng Wang, Dan Xu · 2025 · Neural Information Processing Systems · cited 5 times · Not summarized yet
HyRF combines explicit Gaussians with neural fields to reduce memory overhead while preserving high-quality novel-view synthesis.
Abstract
Recently, 3D Gaussian Splatting (3DGS) has emerged as a powerful alternative to NeRF-based approaches, enabling real-time, high-quality novel view synthesis through explicit, optimizable 3D Gaussians. However, 3DGS suffers from significant memory overhead due to its reliance on per-Gaussian parameters to model view-dependent effects and anisotropic shapes. While recent works propose compressing 3DGS with neural fields, these methods struggle to capture high-frequency spatial variations in Gaussian properties, leading to degraded reconstruction of fine details. We present Hybrid Radiance Fields (HyRF), a novel scene representation that combines the strengths of explicit Gaussians and neural f…
-
F-3DGS: Factorized Coordinates and Representations for 3D Gaussian SplattingXiang-Yu Sun, Joo Chan Lee, Daniel Rho et al. · 2024 · ACM Multimedia · cited 30 times · Not summarized yet
F-3DGS factorizes Gaussian coordinates and representations to reduce the storage required for high-quality 3D scenes.
Abstract
The neural radiance field (NeRF) has made significant strides in representing 3D scenes and synthesizing novel views. Despite its advancements, the high computational costs of NeRF have posed challenges for its deployment in resource-constrained environments and real-time applications. As an alternative to NeRF-like neural rendering methods, 3D Gaussian Splatting (3DGS) offers rapid rendering speeds while maintaining excellent image quality. However, as it represents objects and scenes using a myriad of Gaussians, it requires substantial storage to achieve high-quality representation. To mitigate the storage overhead, we propose Factorized 3D Gaussian Splatting (F-3DGS), a novel approach tha…
-
Color-cued Efficient Densification Method for 3D Gaussian SplattingSieun Kim, Kyungjin Lee, Youngki Lee · 2024 · 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · cited 37 times · Not summarized yet
Color-cued Efficient Densification changes how Gaussians are populated to improve quality with fewer points and smaller models.
Abstract
Many variants of Neural Radiance Fields (NeRF) have been explored in pursuit of high-quality results with reasonable data size and real-time rendering speed. 3D Gaussian Splatting (3DGS) gained popularity due to its ability to render quality images in real-time; however, it still faces challenges with large data sizes. Meanwhile, the densification process of 3DGS plays a large role in deciding the quality and the data size of a model. Hence, it is crucial to devise a densification method that can populate Gaussians efficiently so that quality can be enhanced and fewer Gaussians are used. An efficient densification method that results in fewer Gaussians can also promote efficiency in training…
-
Compact 3D Scene Representation via Self-Organizing Gaussian GridsWieland Morgenstern, Florian Barthel, Anna Hilsmann et al. · 2023 · Computer Vision and Pattern Recognition · cited 226 times · AI summary available
Self-Organizing Gaussian Grids organize Gaussian parameters into a 2D grid to reduce storage without compromising rendering quality.
Abstract
3D Gaussian Splatting has recently emerged as a highly promising technique for modeling of static 3D scenes. In contrast to Neural Radiance Fields, it utilizes efficient rasterization allowing for very fast rendering at high-quality. However, the storage size is significantly higher, which hinders practical deployment, e.g. on resource constrained devices. In this paper, we introduce a compact scene representation organizing the parameters of 3D Gaussian Splatting (3DGS) into a 2D grid with local homogeneity, ensuring a drastic reduction in storage requirements without compromising visual quality during rendering. Central to our idea is the explicit exploitation of perceptual redundancies pr…
-
NeuralGS: Bridging Neural Fields and 3D Gaussian Splatting for Compact 3D RepresentationsZhen-Yu Tang, Chaoran Feng, Xinhua Cheng et al. · 2025 · AAAI Conference on Artificial Intelligence · cited 26 times · Not summarized yet
NeuralGS encodes Gaussian attributes with MLPs and clustering to compress 3DGS scene representations.
Abstract
3D Gaussian Splatting (3DGS) achieves impressive quality and rendering speed, but with millions of 3D Gaussians and significant storage and transmission costs. In this paper, we aim to develop a simple yet effective method called NeuralGS that compresses the original 3DGS into a compact representation. Our observation is that neural fields like NeRF can represent complex 3D scenes with Multi-Layer Perceptron (MLP) neural networks using only a few megabytes. Thus, NeuralGS effectively adopts the neural field representation to encode the attributes of 3D Gaussians with MLPs, only requiring a small storage size even for a large-scale scene. To achieve this, we adopt a clustering strategy and fi…
-
NeRF is a Valuable Assistant for 3D Gaussian SplattingShuangkang Fang, I-Chao Shen, Takeo Igarashi et al. · 2025 · IEEE International Conference on Computer Vision · cited 21 times · Not summarized yet
NeRF-GS jointly optimizes NeRF and 3DGS, using shared spatial information to address initialization and spatial-awareness limitations.
Abstract
We introduce NeRF-GS, a novel framework that jointly optimizes Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS). This framework leverages the inherent continuous spatial representation of NeRF to mitigate several limitations of 3DGS, including sensitivity to Gaussian initialization, limited spatial awareness, and weak inter-Gaussian correlations, thereby enhancing its performance. In NeRF-GS, we revisit the design of 3DGS and progressively align its spatial features with NeRF, enabling both representations to be optimized within the same scene through shared 3D spatial information. We further address the formal distinctions between the two approaches by optimizing residual vect…
Prior work it builds on
-
Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance FieldsJ. Barron, B. Mildenhall, Dor Verbin et al. · 2023 · IEEE International Conference on Computer Vision · cited 947 times · Not summarized yet
Zip-NeRF develops anti-aliased grid-based radiance fields, a grid-based neural representation relevant to the paper’s color-field approach.
Abstract
Neural Radiance Field training can be accelerated through the use of grid-based representations in NeRF’s learned mapping from spatial coordinates to colors and volumetric density. However, these grid-based approaches lack an explicit understanding of scale and therefore often introduce aliasing, usually in the form of jaggies or missing scene content. Anti-aliasing has previously been addressed by mip-NeRF 360, which reasons about sub-volumes along a cone rather than points along a ray, but this approach is not natively compatible with current grid-based techniques. We show how ideas from rendering and signal processing can be used to construct a technique that combines mip-NeRF 360 and gri…
-
Masked Wavelet Representation for Compact Neural Radiance FieldsDaniel Rho, Byeonghyeon Lee, Seungtae Nam et al. · 2022 · Computer Vision and Pattern Recognition · cited 79 times · Not summarized yet
Masked Wavelet Representation compresses neural radiance fields with wavelets and masks rather than compressing Gaussian attributes.
Abstract
Neural radiance fields (NeRF) have demonstrated the potential of coordinate-based neural representation (neural fields or implicit neural representation) in neural rendering. However, using a multi-layer perceptron (MLP) to represent a 3D scene or object requires enormous computational resources and time. There have been recent studies on how to reduce these computational inefficiencies by using additional data structures, such as grids or trees. Despite the promising performance, the explicit data structure necessitates a substantial amount of memory. In this work, we present a method to reduce the size without compromising the advantages of having additional data structures. In detail, we …
-
Direct Voxel Grid Optimization: Super-fast Convergence for Radiance Fields ReconstructionCheng Sun, Min Sun, Hwann-Tzong Chen · 2021 · Computer Vision and Pattern Recognition · cited 1,431 times · Not summarized yet
Direct Voxel Grid Optimization uses density and feature voxel grids for fast radiance-field reconstruction rather than Gaussian compression.
Abstract
We present a super-fast convergence approach to reconstructing the per-scene radiance field from a set of images that capture the scene with known poses. This task, which is often applied to novel view synthesis, is recently revolution-ized by Neural Radiance Field (NeRF) for its state-of-the-art quality and fiexibility. However, NeRF and its variants require a lengthy training time ranging from hours to days for a single scene. In contrast, our approach achieves NeRF-comparable quality and converges rapidly from scratch in less than 15 minutes with a single GPU. We adopt a representation consisting of a density voxel grid for scene geometry and a feature voxel grid with a shallow network fo…
-
RegNeRF: Regularizing Neural Radiance Fields for View Synthesis from Sparse InputsMichael Niemeyer, J. Barron, B. Mildenhall et al. · 2021 · Computer Vision and Pattern Recognition · cited 899 times · Not summarized yet
RegNeRF regularizes novel-view synthesis from sparse inputs, addressing NeRF reconstruction quality rather than Gaussian storage.
Abstract
Neural Radiance Fields (NeRF) have emerged as a powerful representation for the task of novel view synthesis due to their simplicity and state-of-the-art performance. Though NeRF can produce photorealistic renderings of unseen viewpoints when many input views are available, its performance drops significantly when this number is reduced. We observe that the majority of artifacts in sparse input scenarios are caused by errors in the estimated scene geometry, and by divergent behavior at the start of training. We address this by regularizing the geometry and appearance of patches rendered from unobserved viewpoints, and annealing the ray sampling space during training. We additionally use a no…
-
D-NeRF: Neural Radiance Fields for Dynamic ScenesAlbert Pumarola, Enric Corona, G. Pons-Moll et al. · 2020 · Computer Vision and Pattern Recognition · cited 2,165 times · Not summarized yet
D-NeRF extends radiance fields to dynamic scenes, whereas the paper focuses on compacting Gaussian scene representations.
Abstract
Neural rendering techniques combining machine learning with geometric reasoning have arisen as one of the most promising approaches for synthesizing novel views of a scene from a sparse set of images. Among these, stands out the Neural radiance fields (NeRF) [31], which trains a deep network to map 5D input coordinates (representing spatial location and viewing direction) into a volume density and view-dependent emitted radiance. However, despite achieving an unprecedented level of photorealism on the generated images, NeRF is only applicable to static scenes, where the same spatial location can be queried from different images. In this paper we introduce D-NeRF, a method that extends neural…
Papers that cite it
-
Taming 3DGS: High-Quality Radiance Fields with Limited ResourcesSaswat Subhajyoti Mallick, Rahul Goel, Bernhard Kerbl et al. · 2024 · SIGGRAPH Asia Conference Papers · cited 134 times · Not summarized yet
Taming 3DGS reduces resource demands through guided densification, targeting redundant Gaussians and rendering cost.
Abstract
3D Gaussian Splatting (3DGS) has transformed novel-view synthesis with its fast, interpretable, and high-fidelity rendering. However, its resource requirements limit its usability. Especially on constrained devices, training performance degrades quickly and often cannot complete due to excessive memory consumption of the model. The method converges with an indefinite number of Gaussians -- many of them redundant -- making rendering unnecessarily slow and preventing its usage in downstream tasks that expect fixed-size inputs. To address these issues, we tackle the challenges of training and rendering 3DGS models on a budget. We use a guided, purely constructive densification process that stee…
-
CompGS: Smaller and Faster Gaussian Splatting with Vector QuantizationK. Navaneet, Kossar Pourahmadi Meibodi, Soroush Abbasi Koohpayegani et al. · 2023 · European Conference on Computer Vision · cited 190 times · Not summarized yet
CompGS quantizes Gaussian parameters with codebooks, sharing the paper’s use of vector quantization to reduce storage.
Abstract
3D Gaussian Splatting (3DGS) is a new method for modeling and rendering 3D radiance fields that achieves much faster learning and rendering time compared to SOTA NeRF methods. However, it comes with a drawback in the much larger storage demand compared to NeRF methods since it needs to store the parameters for several 3D Gaussians. We notice that many Gaussians may share similar parameters, so we introduce a simple vector quantization method based on K-means to quantize the Gaussian parameters while optimizing them. Then, we store the small codebook along with the index of the code for each Gaussian. We compress the indices further by sorting them and using a method similar to run-length enc…
-
RacerGS: Residual-Autoregressive Context for 3D Gaussian Splatting CompressionWen-Qing Wang, Hui-Min Zeng, Yun Fu · 2026 · Not summarized yet
RacerGS compresses 3DGS with hash-projected context features and autoregressive entropy modeling.
Abstract
3D Gaussian Splatting (3DGS) has emerged as a promising approach for fast, high-quality novel-view rendering. However, its large parameter size remains a bottleneck for storage and transmission. We propose RacerGS, a compact 3DGS compression framework that achieves scene reconstruction at significantly reduced storage sizes with preserved fidelity. To obtain compact and spatially coherent features, we design a hash projection model that projects the anchor hash encodings into compact context features. Additionally, we introduce a FiLM-modulated autoregressive GRU context model that utilizes causal dependencies among anchor feature groups and uses recurrent hidden states to refine entropy-mod…
-
Feed-Forward 3D Gaussian Splatting Compression with Long-Context ModelingZhe-Ning Liu, Rui Song, Yu-Shi Huang et al. · 2025 · arXiv.org · preprint, not peer reviewed · cited 8 times · Not summarized yet
Feed-Forward 3D Gaussian Splatting Compression uses long-range context modeling to reduce 3DGS data size.
Abstract
3D Gaussian Splatting (3DGS) has emerged as a revolutionary 3D representation. However, its substantial data size poses a major barrier to widespread adoption. While feed-forward 3DGS compression offers a practical alternative to costly per-scene per-train compressors, existing methods struggle to model long-range spatial dependencies, due to the limited receptive field of transform coding networks and the inadequate context capacity in entropy models. In this work, we propose a novel feed-forward 3DGS compression framework that effectively models long-range correlations to enable highly compact and generalizable 3D representations. Central to our approach is a large-scale context structure …
-
A Survey on 3D Gaussian Splatting Applications: Segmentation, Editing, and GenerationShu-Ting He, P. Ji, Yi-Tong Yang et al. · 2025 · IEEE Transactions on Pattern Analysis and Machine Intelligence · cited 32 times · Not summarized yet
The survey reviews 3DGS applications and related research, including reconstruction preliminaries and NeRF-based work.
Abstract
In the context of novel view synthesis, 3D Gaussian Splatting (3DGS) has recently emerged as an efficient and competitive counterpart to Neural Radiance Field (NeRF), enabling high-fidelity photorealistic rendering in real time. Beyond novel view synthesis, the explicit and compact nature of 3DGS enables a wide range of downstream applications that require geometric and semantic understanding. This survey provides a comprehensive overview of recent progress in 3DGS applications. It first reviews the reconstruction preliminaries of 3DGS, followed by the problem formulation, 2D foundation models, and related NeRF-based research areas that inform downstream 3DGS applications. We then categorize…
Go further
How are these papers found?
We compare the title and abstract of this paper with those of every paper summarized on this site, and with the papers Semantic Scholar recommends, those it cites and those citing it (for a much-cited paper, its most recent citations and, from OpenAlex, the most cited ones). Among the prior work and the follow-ups, the papers that use each other in depth, a method it uses and the much-cited follow-ups come a little higher. The levels (very close, close, related) come from the similarity of the texts, not from a judgment of quality; * marks a paper without an abstract, compared by its title. The sentence under a paper is written by an AI from the two abstracts: check it in the papers. A paper marked not relevant by three readers leaves the list.
Updated on 4 November 2026.