Two-View Accumulation as the Primary Training Lever for Hybrid-Capture Gaussian Splatting: A Variance-Decomposition View of When Gradient Surgery Helps

• URL: http://arxiv.org/abs/2605.00052v1
• Date: Wed, 29 Apr 2026 17:45:51 GMT
• Title: Two-View Accumulation as the Primary Training Lever for Hybrid-Capture Gaussian Splatting: A Variance-Decomposition View of When Gradient Surgery Helps
• Abstract summary: Hybrid-capture novel view synthesis combines substantially different camera views.<n>Standard 3DGS is trained for 30K iterations with one rendered view per step.<n>We propose a variance-decomposition framework that predicts and explains this finding.
• Score: 7.6889618752994595
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Hybrid-capture novel view synthesis combines images at substantially different camera distances (e.g., aerial drone and ground-level views). Standard 3D Gaussian Splatting (3DGS), trained for 30K iterations with one rendered view per optimizer step, under-fits the minority regime by 1-3 dB on five hybrid-capture benchmarks. We isolate the lever that closes this gap. Among compute-matched alternatives -- vanilla 60K iterations, magnitude corrections (GradNorm), direction-aware near/far gradient surgery, projective preconditioning, confidence-gated sample-level surgery, and a random two-view-per-step control -- the simplest structural change wins: rendering two views per optimizer step. The pairing rule (geometry-defined near/far, random, or active loss-disparity) does not change PSNR beyond seed variance on any of the five scenes; the structural change of having two views per step does. We propose a variance-decomposition framework that predicts and explains this finding: under bimodal camera regimes, between-regime gradient variance turns out to be small relative to within-regime variance in 3DGS, so structured and random pairings are variance-equivalent in expectation, and the variance halving from two-view accumulation itself is the dominant effect. We verify the framework on five scenes whose camera-altitude bimodality coefficients span [0.55, 1.00], and we report the negative result that direction-aware projection, magnitude correction, confidence gating, and an active loss-disparity pairing all fall within seed variance of random two-view pairing. The two-view structural lever transfers cleanly to the Scaffold-GS and Pixel-GS backbones. We position this work as an honest characterization of which training-side axes do and do not move PSNR for hybrid-capture 3DGS, together with the framework that explains why.

Related papers

• 3D Digital Twin Visualization of Multiclass GRF-Based Gait Disorder Classification [3.4540258577108776]
  We propose an integrated framework for classifying healthy gait and multiple musculoskeletal impairment groups.<n>The signals were normalized over the stance phase and standardized using training-set statistics.<n>The model achieved a validation accuracy of 99.00% and a test accuracy of 90.07% under a session-level split.
  arXiv  Detail & Related papers  (2026-09-11T05:04:08Z)
• NCGR: Noise-Conditional Gated Rectification for Camera Extrinsic Perturbations in BEV 3D Object Detection [15.441348570437194]
  Camera-based bird's-eye-view (BEV) 3D detection typically assumes accurate and fixed camera extrinsics.<n>In detectors using spatial cross-attention (SCA), extrinsic perturbations displace the image-plane projections of BEV reference points.<n>Noise-Conditional Gated Rectification (NCGR) is proposed to compensate for projection errors without explicitly estimating a full six-degree-of-freedom extrinsic correction.
  arXiv  Detail & Related papers  (2026-08-04T16:26:57Z)
• GrainGS: Gradient-Decoupled Gaussian Splatting for Efficient Dynamic Novel View Synthesis [94.20462261657525]
  GrainGS is a dynamic framework that combines a hierarchical anchor scaffold with per-Gaussian deformation.<n>It achieves high reconstruction quality, real-time novel view synthesis, and compact storage.
  arXiv  Detail & Related papers  (2026-07-23T15:52:35Z)
• DisPOSE: Projected Polystochastic Diffusion for Self-Supervised Multi-View 3D Human Pose Estimation [58.47973015036709]
  DisPOSE is a self-supervised framework that approximates the inherently discrete multi-view person-assignment problem.<n>By employing differentiable Sinkhorn projections during denoising, our model learns to guide solutions toward valid and feasible assignments.<n>The proposed approach outperforms current state-of-the-art self-supervised methods on standard datasets.
  arXiv  Detail & Related papers  (2026-06-05T16:14:39Z)
• iGaussian: Real-Time Camera Pose Estimation via Feed-Forward 3D Gaussian Splatting Inversion [62.09575122593993]
  iGaussian is a two-stage feed-forward framework that achieves real-time camera pose estimation through direct 3D Gaussian inversion.<n> Experimental results on the NeRF Synthetic, Mip-NeRF 360, and T&T+DB datasets demonstrate a significant performance improvement over previous methods.
  arXiv  Detail & Related papers  (2025-11-18T05:22:22Z)
• SPFSplatV2: Efficient Self-Supervised Pose-Free 3D Gaussian Splatting from Sparse Views [18.814209805277503]
  SPFSplatV2, an efficient feed-forward framework for 3D Gaussian splatting from sparse multi-view images, is presented.<n>Method achieves state-of-the-art performance in both in-domain and out-of-domain novel view synthesis.
  arXiv  Detail & Related papers  (2025-09-21T21:37:56Z)
• Quantifying and Alleviating Co-Adaptation in Sparse-View 3D Gaussian Splatting [39.014517076251934]
  3D Gaussian Splatting (3DGS) has demonstrated impressive performance in novel view synthesis under dense-view settings.<n>In sparse-view scenarios, despite the realistic renderings in training views, 3DGS occasionally manifests appearance artifacts in novel views.<n>This paper investigates the appearance artifacts in sparse-view 3DGS and uncovers a core limitation of current approaches.
  arXiv  Detail & Related papers  (2025-08-18T08:34:49Z)
• PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting [54.7468067660037]
  PF3plat sets a new state-of-the-art across all benchmarks, supported by comprehensive ablation studies validating our design choices.<n>Our framework capitalizes on fast speed, scalability, and high-quality 3D reconstruction and view synthesis capabilities of 3DGS.
  arXiv  Detail & Related papers  (2024-10-29T15:28:15Z)
• S^2Former-OR: Single-Stage Bi-Modal Transformer for Scene Graph Generation in OR [50.435592120607815]
  Scene graph generation (SGG) of surgical procedures is crucial in enhancing holistically cognitive intelligence in the operating room (OR) Previous works have primarily relied on multi-stage learning, where the generated semantic scene graphs depend on intermediate processes with pose estimation and object detection. In this study, we introduce a novel single-stage bi-modal transformer framework for SGG in the OR, termed S2Former-OR.
  arXiv  Detail & Related papers  (2024-02-22T11:40:49Z)
• Multi-body SE(3) Equivariance for Unsupervised Rigid Segmentation and Motion Estimation [49.56131393810713]
  We present an SE(3) equivariant architecture and a training strategy to tackle this task in an unsupervised manner. Our method excels in both model performance and computational efficiency, with only 0.25M parameters and 0.92G FLOPs.
  arXiv  Detail & Related papers  (2023-06-08T22:55:32Z)
• Out-of-Domain Human Mesh Reconstruction via Dynamic Bilevel Online Adaptation [87.85851771425325]
  We consider a new problem of adapting a human mesh reconstruction model to out-of-domain streaming videos. We tackle this problem through online adaptation, gradually correcting the model bias during testing. We propose the Dynamic Bilevel Online Adaptation algorithm (DynaBOA)
  arXiv  Detail & Related papers  (2021-11-07T07:23:24Z)

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.