Improving Continual Learning for Gaussian Splatting based Environments Reconstruction on Commercial Off-the-Shelf Edge Devices

• URL: http://arxiv.org/abs/2603.08499v1
• Date: Mon, 09 Mar 2026 15:32:41 GMT
• Title: Improving Continual Learning for Gaussian Splatting based Environments Reconstruction on Commercial Off-the-Shelf Edge Devices
• Abstract summary: Novel view synthesis (NVS) is increasingly relevant for edge robotics, where compact and incrementally updatable 3D scene models are needed.<n>We present a precision-adaptive framework that enables VBGS training on resource-constrained hardware without altering its variational formulation.<n>We also enable for the first time NVS training on a commercial embedded platform, the Jetson Orin Nano, reducing per-frame latency by 19x compared to 3DGS.
• Score: 2.48245150714679
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Novel view synthesis (NVS) is increasingly relevant for edge robotics, where compact and incrementally updatable 3D scene models are needed for SLAM, navigation, and inspection under tight memory and latency budgets. Variational Bayesian Gaussian Splatting (VBGS) enables replay-free continual updates for the 3DGS algorithm by maintaining a probabilistic scene model, but its high-precision computations and large intermediate tensors make on-device training impractical. We present a precision-adaptive optimization framework that enables VBGS training on resource-constrained hardware without altering its variational formulation. We (i) profile VBGS to identify memory/latency hotspots, (ii) fuse memory-dominant kernels to reduce materialized intermediate tensors, and (iii) automatically assign operation-level precisions via a mixed-precision search with bounded relative error. Across the Blender, Habitat, and Replica datasets, our optimised pipeline reduces peak memory from 9.44 GB to 1.11 GB and training time from ~234 min to ~61 min on an A5000 GPU, while preserving (and in some cases improving) reconstruction quality of the state-of-the-art VBGS baseline. We also enable for the first time NVS training on a commercial embedded platform, the Jetson Orin Nano, reducing per-frame latency by 19x compared to 3DGS.

Related papers

• CVT-GS: Learning to Simplify 3D Gaussian Splatting with Centroidal Voronoi Tessellation [42.60357708039613]
  3D Gaussian Splatting (3DGS) has emerged as a powerful representation for real-time novel view synthesis.<n> rendering high-fidelity scenes often relies on a massive number of Gaussian primitives.<n>Existing simplification techniques are largely intrusive, requiring training-time pruning, architectural modifications, or computationally expensive per-scene fine-tuning.<n>We propose Centro-GS, a novel optimization-free post-hoc simplification framework that directly compresses trained 3DGS scenes without sacrificing visual fidelity.
  arXiv  Detail & Related papers  (2026-09-08T13:26:32Z)
• TurboGS: Accelerating 3D Gaussian Splatting via Error-Guided Sparse Pixel Sampling and Optimization [54.179316928687435]
  TurboGS is an error-guided training framework that accelerates 3DGS by concentrating optimization on perceptually informative pixels.<n>Experiments on standard benchmarks demonstrate that TurboGS can deliver on par or superior rendering quality within 100 seconds on a single 5090 GPU card.
  arXiv  Detail & Related papers  (2026-06-14T16:59:48Z)
• Gaussians on a Diet: High-Quality Memory-Bounded 3D Gaussian Splatting Training [23.288643132782216]
  3D Gaussian Splatting (3DGS) has revolutionized novel view synthesis with high-quality rendering through continuous aggregations of millions of 3D Gaussian primitives.<n>It suffers from a substantial memory footprint, particularly during training due to uncontrolled densification.<n>We propose a memory-bounded training framework that dynamically optimize Gaussians through iterative growth and pruning.
  arXiv  Detail & Related papers  (2026-04-21T23:04:03Z)
• ImprovedGS+: A High-Performance C++/CUDA Re-Implementation Strategy for 3D Gaussian Splatting [0.0]
  We propose ImprovedGS+, a high-performance, low-level reinvention of the ImprovedGS strategy.<n>By transitioning from high-level Python logic to hardware-optimized C++/CUDA kernels, we achieve a significant reduction in host-device synchronization and training latency.
  arXiv  Detail & Related papers  (2026-03-09T17:38:27Z)
• Speedy Deformable 3D Gaussian Splatting: Fast Rendering and Compression of Dynamic Scenes [57.69608119350651]
  Recent extensions of 3D Gaussian Splatting (3DGS) to dynamic scenes achieve high-quality novel view synthesis by using neural networks to predict the time-varying deformation of each Gaussian.<n>However, performing per-Gaussian neural inference at every frame poses a significant bottleneck, limiting rendering speed and increasing memory and compute requirements.<n>We present Speedy Deformable 3D Gaussian Splatting (SpeeDe3DGS), a general pipeline for accelerating the rendering speed of dynamic 3DGS and 4DGS representations by reducing neural inference through two complementary techniques.
  arXiv  Detail & Related papers  (2025-06-09T16:30:48Z)
• Steepest Descent Density Control for Compact 3D Gaussian Splatting [72.54055499344052]
  3D Gaussian Splatting (3DGS) has emerged as a powerful real-time, high-resolution novel view.<n>We propose a theoretical framework that demystifies and improves density control in 3DGS.<n>We introduce SteepGS, incorporating steepest density control, a principled strategy that minimizes loss while maintaining a compact point cloud.
  arXiv  Detail & Related papers  (2025-05-08T18:41:38Z)
• Optimized 3D Gaussian Splatting using Coarse-to-Fine Image Frequency Modulation [22.041183774936144]
  We propose Opti3DGS, a novel frequency-modulated coarse-to-fine optimization framework.<n>We show that our method integrates seamlessly with many 3DGS-based techniques.<n>We also show that Opti3DGS inherently produces a level-of-detail scene representation at no extra cost.
  arXiv  Detail & Related papers  (2025-03-18T17:49:01Z)
• LiteGS: A High-performance Framework to Train 3DGS in Subminutes via System and Algorithm Codesign [6.7123686385076615]
  3D Gaussian Splatting (3DGS) has emerged as promising alternative in 3D representation, but it still suffers from high training cost.<n>This paper introduces LiteGS, a high performance computation framework that systematically optimize the 3DGS training pipeline.<n> Experimental results demonstrate that LiteGS accelerates the original 3DGS training by up to 13.4x with comparable or superior quality.
  arXiv  Detail & Related papers  (2025-03-03T05:52:02Z)
• Taming 3DGS: High-Quality Radiance Fields with Limited Resources [50.92437599516609]
  3D Gaussian Splatting (3DGS) has transformed novel-view synthesis with its fast, interpretable, and high-fidelity rendering. We tackle the challenges of training and rendering 3DGS models on a budget. We derive faster, numerically equivalent solutions for gradient computation and attribute updates.
  arXiv  Detail & Related papers  (2024-06-21T20:44:23Z)
• PUP 3D-GS: Principled Uncertainty Pruning for 3D Gaussian Splatting [59.277480452459315]
  We propose a principled sensitivity pruning score that preserves visual fidelity and foreground details at significantly higher compression ratios.<n>We also propose a multi-round prune-refine pipeline that can be applied to any pretrained 3D-GS model without changing its training pipeline.
  arXiv  Detail & Related papers  (2024-06-14T17:53:55Z)
• LP-3DGS: Learning to Prune 3D Gaussian Splatting [71.97762528812187]
  We propose learning-to-prune 3DGS, where a trainable binary mask is applied to the importance score that can find optimal pruning ratio automatically. Experiments have shown that LP-3DGS consistently produces a good balance that is both efficient and high quality.
  arXiv  Detail & Related papers  (2024-05-29T05:58:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.