mlsanityData to 7 Oct 2026

LGS: A Light-weight 4D Gaussian Splatting for Efficient Surgical Scene Reconstruction

arXiv 2406.16073 ↗First posted 23 Jun 2024cs.CVcited by 21

Not on the map. Why?

Abstract (arXiv)

The advent of 3D Gaussian Splatting (3D-GS) techniques and their dynamic scene modeling variants, 4D-GS, offers promising prospects for real-time rendering of dynamic surgical scenarios. However, the prerequisite for modeling dynamic scenes by a large number of Gaussian units, the high-dimensional Gaussian attributes and the high-resolution deformation fields, all lead to serve storage issues that hinder real-time rendering in resource-limited surgical equipment. To surmount these limitations, we introduce a Lightweight 4D Gaussian Splatting framework (LGS) that can liberate the efficiency bottlenecks of both rendering and storage for dynamic endoscopic reconstruction. Specifically, to minimize the redundancy of Gaussian quantities, we propose Deformation-Aware Pruning by gauging the impact of each Gaussian on deformation. Concurrently, to reduce the redundancy of Gaussian attributes, we simplify the representation of textures and lighting in non-crucial areas by pruning the dimensions of Gaussian attributes. We further resolve the feature field redundancy caused by the high resolution of 4D neural spatiotemporal encoder for modeling dynamic scenes via a 4D feature field condensation. Experiments on public benchmarks demonstrate efficacy of LGS in terms of a compression rate exceeding 9 times while maintaining the pleasing visual quality and real-time rendering efficiency. LGS confirms a substantial step towards its application in robotic surgical services.

Key facts. Cited by 21 papers in the map (data through 7 Oct 2026). Builds on 33 of its 54 references that are in the map. First posted on arXiv 23 Jun 2024.

Builds on

33 of 54 references.

References · 33

  1. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  2. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,141
  3. 2021Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fieldscited by 1,957
  4. 20234D Gaussian Splatting for Real-Time Dynamic Scene Renderingcited by 1,144
  5. 2023LightGaussian: Unbounded 3D Gaussian Compression with 15x Reduction and 200+ FPScited by 420
  6. 2023Compact 3D Gaussian Representation for Radiance Fieldcited by 365
  7. 2019Differentiable Surface Splatting for Point-based Geometry Processingcited by 238
  8. 2024U-KAN Makes Strong Backbone for Medical Image Segmentation and Generationcited by 176
  9. 2023EfficientViT: Memory Efficient Vision Transformer with Cascaded Group Attentioncited by 172
  10. 2022Neural Rendering for Stereo 3D Reconstruction of Deformable Tissues in Robotic Surgerycited by 122
  11. 2021Stereo Correspondence and Reconstruction of Endoscopic Data Challengecited by 101
  12. 2022SIGMA: Semantic-complete Graph Matching for Domain Adaptive Object Detectioncited by 74
21 more
  1. 2023EndoSurf: Neural Surface Reconstruction of Deformable Tissues with Stereo Endoscope Videoscited by 68
  2. 2024Endora: Video Generation Models as Endoscopy Simulatorscited by 53
  3. 2020Few-shot Medical Image Segmentation using a Global Correlation Network with Discriminative Embeddingcited by 48
  4. 2019SuPer: A Surgical Perception Framework for Endoscopic Tissue Manipulation with Surgical Roboticscited by 46
  5. 2023Neural LerPlane Representations for Fast 4D Reconstruction of Deformable Tissuescited by 44
  6. 2024EndoGS: Deformable Endoscopic Tissues Reconstruction with Gaussian Splattingcited by 33
  7. 2022StegaNeRF: Embedding Invisible Information within Neural Radiance Fieldscited by 32
  8. 2019SurfelWarp: Efficient Non-Volumetric Single View Dynamic Reconstructioncited by 31
  9. 2020SuPer Deep: A Surgical Perception Framework for Robotic Tissue Manipulation using Deep Learning for Feature Extractioncited by 28
  10. 2022Assessing generalisability of deep learning-based polyp detection and segmentation methods through a computer vision challengecited by 27
  11. 2023LightNeuS: Neural Surface Reconstruction in Endoscopy using Illumination Declinecited by 24
  12. 2020Dynamic Reconstruction of Deformable Soft-tissue with Stereo Scope in Minimal Invasive Surgerycited by 23
  13. 2022Knowledge Condensation Distillationcited by 20
  14. 2023Learning to Estimate 6DoF Pose from Limited Data: A Few-Shot, Generalizable Approach using RGB Imagescited by 18
  15. 2022BREAK: Bronchi Reconstruction by gEodesic transformation And sKeleton embeddingcited by 16
  16. 2016ORBSLAM-based Endoscope Tracking and 3D Reconstructioncited by 15
  17. 2020Generator Versus Segmentor: Pseudo-healthy Synthesiscited by 12
  18. 2022AFSC: Adaptive Fourier Space Compression for Anomaly Detectioncited by 12
  19. 2021Unsupervised Anomaly Segmentation using Image-Semantic Cycle Translationcited by 10
  20. 2022Tracking monocular camera pose and deformation for SLAM inside the human bodycited by 10
  21. 2024EndoGaussians: Single View Dynamic Gaussian Splatting for Deformable Endoscopic Tissues Reconstructioncited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 7; 2025: 12; 2026: 2.

21 papers.

Sort:

  1. 2024EndoSparse: Real-Time Sparse View Synthesis of Endoscopic Scenes using Gaussian Splattingcited by 16
  2. 2025VoxDet: Rethinking 3D Semantic Occupancy Prediction as Dense Object Detectioncited by 12
  3. 2025Perceptual-GS: Scene-adaptive Perceptual Densification for Gaussian Splattingcited by 12
  4. 2025Dynamic Scene Reconstruction: Recent Advance in Real-time Rendering and Streamingcited by 8
  5. 2024GTP-4o: Modality-prompted Heterogeneous Graph Learning for Omni-modal Biomedical Representationcited by 7
  6. 2024A Review of 3D Reconstruction Techniques for Deformable Tissues in Robotic Surgerycited by 7
  7. 2024Multi-Layer Gaussian Splatting for Immersive Anatomy Visualizationcited by 7
  8. 2025Deformable Gaussian Splatting for Efficient and High-Fidelity Reconstruction of Surgical Scenescited by 6
  9. 2024When 3D Partial Points Meets SAM: Tooth Point Cloud Segmentation with Sparse Labelscited by 5
  10. 2025Instrument-Splatting: Controllable Photorealistic Reconstruction of Surgical Instruments Using Gaussian Splattingcited by 5
  11. 2025SpeeDe3DGS: Speedy Deformable 3D Gaussian Splatting with Temporal Pruning and Motion Groupingcited by 4
  12. 2025BridgeSplat: Bidirectionally Coupled CT and Non-Rigid Gaussian Splatting for Deformable Intraoperative Surgical Navigationcited by 4
9 more
  1. 2025ConcealGS: Concealing Invisible Copyright Information in 3D Gaussian Splattingcited by 3
  2. 2024Real-Time Spatio-Temporal Reconstruction of Dynamic Endoscopic Scenes with 4D Gaussian Splattingcited by 2
  3. 2025Moving Light Adaptive Colonoscopy Reconstruction via Illumination-Attenuation-Aware 3D Gaussian Splattingcited by 1
  4. 20264D Monocular Surgical Reconstruction under Arbitrary Camera Motionscited by 1
  5. 2024DaRePlane: Direction-aware Representations for Dynamic Scene Reconstructioncited by 0
  6. 2025Feature-EndoGaussian: Feature Distilled Gaussian Splatting in Surgical Deformable Scene Reconstructioncited by 0
  7. 2025MetaScope: Optics-Driven Neural Network for Ultra-Micro Metalens Endoscopycited by 0
  8. 2025Effective Gaussian Management for High-fidelity Scene Reconstructioncited by 0
  9. 2026Monocular Endoscopic Tissue 3D Reconstruction with Multi-Level Geometry Regularizationcited by 0