Mind the Gap: Standard 3DGS Evaluation Primarily Measures Near-Trajectory Interpolation

• URL: http://arxiv.org/abs/2607.01556v1
• Date: Thu, 02 Jul 2026 00:24:15 GMT
• Title: Mind the Gap: Standard 3DGS Evaluation Primarily Measures Near-Trajectory Interpolation
• Abstract summary: Standard 3D Gaussian Splatting evaluation holds out every N-th frame -- but these frames have trained neighbors on both sides, so the metric measures near-trajectory rather than spatial generalization.<n>We introduce a fair matched-count protocol that isolates this effect.<n>Both arms train on the same number of images and differ only in whether the holdout is spread evenly (interpolation) or forms a contiguous spatial sector (extrapolation)
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Standard MipNeRF360-style 3D Gaussian Splatting (3DGS) evaluation holds out every N-th frame -- but these frames have trained neighbors on both sides, so the metric measures near-trajectory interpolation rather than spatial generalization. We introduce a fair matched-count protocol that isolates this effect: both arms train on the same number of images and differ only in whether the holdout is spread evenly (interpolation) or forms a contiguous spatial sector (extrapolation). Our primary finding is a large, consistent interpolation-extrapolation gap of 3~12dB -- several times the differences typically reported between competing methods. The gap is robust to training noise, is in two cases large enough to flip a method ranking under multi-seed confirmation, and -- crucially -- persists across three representation families, including a non-Gaussian volumetric neural radiance field (NeRF), so it reflects spatial coverage rather than any one representation. Diagnostically, it is dominated by a diffuse/geometry-proxy component and tracks each view's angular distance to its nearest training view, a zero-cost signal that also guides capture planning; loss-side regularization yields only marginal gains. Standard holdouts remain useful for near-trajectory rendering but should not, alone, be read as evidence of spatial generalization. Prior work notes protocol sensitivity; ours is, to our knowledge, the first to combine matched-count paired holdout, cross-representation quantification, and a diagnostic analysis Table 1. We describe a spatial-holdout benchmark toolkit with standardized splits and baselines for 16 scenes, which we are preparing for public release.

Related papers

• When Does Consensus Mean Correctness? Measuring the Agreement-Accuracy Coupling with Semantics-Preserving Re-Rendering [0.8594140167290097]
  A model's agreement across perturbed inputs is used as a label-free reliability signal.<n>Rerendering beats resampling on both accuracy and reliability.<n>Fine-tuning on the model's own cross-render consensus inverts.
  arXiv  Detail & Related papers  (2026-08-06T07:08:05Z)
• FoundationGeo: Learning Spatial Pixel-Wise Fields for Monocular Metric Geometry [52.232826445242644]
  FoundationGeo bridges relative and metric prediction via spatial calibration and principled data design.<n>Stage 1 learns a high-fidelity, affine-invariant geometry model by initializing with DINOv3.<n>Stage 2 moves beyond global scaling by introducing pixel-wise calibration fields for metric estimation.
  arXiv  Detail & Related papers  (2026-07-13T14:10:01Z)
• Topology-Driven Transferability Estimation for 3D Medical Vision Foundation Models [24.834339924064505]
  We propose a voxel-wise framework for dense prediction tasks in 3D medical imaging.<n>Our framework estimates transferability directly from the alignment between the sparse 1-skeleton geometric graph of dense features and semantic labels.<n>Our framework achieves state-of-the-art transferability estimation with an average weighted Kendall (outperforming by 0.36) while accelerating evaluation by 56 times.
  arXiv  Detail & Related papers  (2026-07-05T09:45:18Z)
• From Sparse and Imperfect 2D Anchors to Consistent 3D Gaussian Street Scenes: Support-Aware Appearance [13.51195065028002]
  Image priors can synthesize target conditions for 3D Gaussian street scenes, but independently edited views do not define a coherent 3D target.<n>To address this gap, teacher-relative appearance residual distillation is introduced for appearance baking.<n>A structured space for frequency decomposition, confidence estimation, and primitive-level lifting is formed by residuals between teacher anchors and original renders.
  arXiv  Detail & Related papers  (2026-06-24T16:26:53Z)
• 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)
• GLINT: Sparsely Gated Vision-Language Alignment for Fine-Grained Radiology Representations [1.2679579476597642]
  We present GLINT (Gated Language-Image alignmeNT), a framework that explicitly models this sparse correspondence.<n> GLINT enables zero-shot classification, grounding, and segmentation from free-text queries.<n>It is the first to demonstrate zero-shot segmentation on 3D CT volumes without mask supervision.
  arXiv  Detail & Related papers  (2026-06-02T05:40:51Z)
• Beyond Euclidean Prototypes: Spectral Disentanglement and Geodesic Matching for Few-Shot Medical Image Segmentation [14.754071441705841]
  Few-Shot Medical Image aims to delineate novel anatomical targets from one or a few annotated support images.<n>Current prototype-based methods are bottlenecked by two coupled limitations.<n>We propose Spectral-Geodesic Prototype Network (SGP-Net), built around a Spectral-Geodesic Prototype Module.
  arXiv  Detail & Related papers  (2026-05-18T06:10:41Z)
• Faster 3D Gaussian Splatting Convergence via Structure-Aware Densification [52.517763621139956]
  3D Gaussian Splatting has emerged as a powerful scene representation for real-time novel-view synthesis.<n>Standard adaptive density control relies on screen-space positional gradients.<n>We present a structure-aware densification framework for 3D Gaussian Splatting.
  arXiv  Detail & Related papers  (2026-04-30T15:37:20Z)
• Robust Scene Coordinate Regression via Geometrically-Consistent Global Descriptors [52.57327385675752]
  We propose an aggregator module that learns global descriptors consistent with both geometrical structure and visual similarity.<n>This corrects erroneous associations caused by unreliable overlap scores.<n>Experiments on challenging benchmarks show substantial localization gains in large-scale environments.
  arXiv  Detail & Related papers  (2025-12-19T04:24:03Z)
• OpenInsGaussian: Open-vocabulary Instance Gaussian Segmentation with Context-aware Cross-view Fusion [89.98812408058336]
  We introduce textbfOpenInsGaussian, an textbfOpen-vocabulary textbfInstance textbfGaussian segmentation framework with Context-aware Cross-view Fusion.<n>OpenInsGaussian achieves state-of-the-art results in open-vocabulary 3D Gaussian segmentation, outperforming existing baselines by a large margin.
  arXiv  Detail & Related papers  (2025-10-21T03:24:12Z)
• PAUL: Uncertainty-Guided Partition and Augmentation for Robust Cross-View Geo-Localization under Noisy Correspondence [16.322806051944227]
  Cross-view geo-localization is a critical task for UAV navigation, event detection, and aerial surveying.<n>Most existing approaches embed multi-modal data into a joint feature space to maximize the similarity of paired images.<n>These methods typically assume perfect alignment of image pairs during training, which rarely holds true in real-world scenarios.
  arXiv  Detail & Related papers  (2025-08-27T17:21:22Z)
• Keep Your Friends Close & Enemies Farther: Debiasing Contrastive Learning with Spatial Priors in 3D Radiology Images [11.251405818285331]
  We propose a 3D contrastive framework (Spade) that leverages extracted correspondences to select more effective positive & negative samples for representation learning. Compared to recent state-of-the-art approaches, Spade shows notable improvements on three downstream segmentation tasks.
  arXiv  Detail & Related papers  (2022-11-16T03:36:06Z)
• Learning to Estimate Hidden Motions with Global Motion Aggregation [71.12650817490318]
  Occlusions pose a significant challenge to optical flow algorithms that rely on local evidences. We introduce a global motion aggregation module to find long-range dependencies between pixels in the first image. We demonstrate that the optical flow estimates in the occluded regions can be significantly improved without damaging the performance in non-occluded regions.
  arXiv  Detail & Related papers  (2021-04-06T10:32:03Z)
• Wide-Depth-Range 6D Object Pose Estimation in Space [124.94794113264194]
  6D pose estimation in space poses unique challenges that are not commonly encountered in the terrestrial setting. One of the most striking differences is the lack of atmospheric scattering, allowing objects to be visible from a great distance. We propose a single-stage hierarchical end-to-end trainable network that is more robust to scale variations.
  arXiv  Detail & Related papers  (2021-04-01T08:39:26Z)

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.