Variance Reduction on the Camera Axis: Multi-View Score Distillation for 3D

• URL: http://arxiv.org/abs/2606.29964v1
• Date: Mon, 29 Jun 2026 08:41:42 GMT
• Title: Variance Reduction on the Camera Axis: Multi-View Score Distillation for 3D
• Abstract summary: Score distillation turns a pretrained 2D diffusion model into a 3D generator, but the per-step gradient is estimated from a single randomly chosen view.<n>Prior work addresses this by retraining the diffusion prior on multi-view data.<n>We instead isolate the sampling axis.
• Score: 0.30586855806896046
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Score distillation turns a pretrained 2D diffusion model into a 3D generator, but the per-step gradient is estimated from a single randomly chosen view: it is high-variance and blind to global shape consistency. Prior work addresses this by retraining the diffusion prior on multi-view data; this improves consistency but makes the sampling contribution inseparable from prior quality. We instead isolate the sampling axis. The per-step gradient is one noisy sample of an expectation over views; aggregating K samples per step at a fixed total UNet budget reduces variance without touching the prior. We introduce Multi-View Aggregated Score Distillation (MV-SDI), which aggregates gradients from K views per step via gradient accumulation, keeping peak memory unchanged and the 2D prior frozen, and draws views as antithetic antipodal pairs, a prior-independent geometric property, for balanced angular coverage. At a fixed 10,000-UNet-call budget, K=2 raises CLIP R-Precision from 74.8% to 83.8% and CLIP score from 0.297 to 0.312, with consistent gains on HPSv2 and ImageReward and a 0.0% divergence rate on the 43-prompt benchmark; optimization steps halve as a consequence. K=4 gives a fourfold step reduction at R-Precision 86.9% and CLIP 0.307, still well above the single-view baseline on every alignment metric. MV-SDI is compatible with gradient-based score-distillation pipelines, including Score Distillation via Inversion, and requires no retraining and no multi-view data.

Related papers

• Spectral Gradient Orthogonalization Improves Differentially Private Training at Scale [17.610398782149293]
  In vision models, where spatial correlation concentrates gradient energy into a low-rank subspace, most of this noise falls in directions that carry little signal.<n> Spectral gradient orthogonalization via polar decomposition is introduced as a post-processing step.<n>It improves accuracy only when the per-direction spectral signal-to-noise ratio (SNR) suffices for singular vector recovery.
  arXiv  Detail & Related papers  (2026-08-18T06:35:26Z)
• Soft-Constrained Optimization of Latent Space in Variational Autoencoders [15.27508597551075]
  We formulate VAE training as a soft-constrained optimization problem that addresses both.<n>We find that low-entropy discrete factors tend to merge into a single latent variable, whereas high-entropy continuous factors are distributed across several.<n>We also find that low-entropy discrete factors tend to merge into a single latent variable, whereas high-entropy continuous factors are distributed across several.
  arXiv  Detail & Related papers  (2026-07-26T16:52:35Z)
• Surflo: Consistent 3D Surface Flow Model with Global State [57.57234680235885]
  Surflo decodes oriented 3D surface points by independently transporting them from noise onto the surface via flow matching.<n>To suppress the local inconsistencies inherent to independent per-point decoding, an inference-time guidance term correlates nearby points by injecting a photometric gradient.<n>Surflo matches or surpasses feed-forward baselines on surface metrics, runs an order of magnitude faster than optimization-based methods that require hundreds of views, and is the only feed-forward approach to combine a global latent with arbitrary-resolution decoding.
  arXiv  Detail & Related papers  (2026-06-11T17:48:38Z)
• AnyFlow: Any-Step Video Diffusion Model with On-Policy Flow Map Distillation [70.80207977907908]
  We introduce AnyFlow, the first any-step video diffusion distillation framework based on flow maps.<n>We show that AnyFlow achieves performance matches or surpasses consistency-based counterparts in the few-step regime.
  arXiv  Detail & Related papers  (2026-05-13T16:06:34Z)
• Two-View Accumulation as the Primary Training Lever for Hybrid-Capture Gaussian Splatting: A Variance-Decomposition View of When Gradient Surgery Helps [7.6889618752994595]
  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.
  arXiv  Detail & Related papers  (2026-04-29T17:45:51Z)
• FSampler: Training Free Acceleration of Diffusion Sampling via Epsilon Extrapolation [0.0]
  FSampler is a training free, sampler execution layer that accelerates diffusion sampling by reducing the number of function evaluations (NFE)<n>FSampler maintains a short history of denoising signals from recent real model calls and extrapolates the next epsilon using finite difference predictors.<n> operating at the sampler level, FSampler integrates with Euler/DDIM, DPM++ 2M/2S, LMS/AB2, and RES family exponential multistep methods.
  arXiv  Detail & Related papers  (2025-11-12T10:21:25Z)
• Environment-Aware Indoor LoRaWAN Path Loss: Parametric Regression Comparisons, Shadow Fading, and Calibrated Fade Margins [3.776919981139063]
  Indoor LoRaWAN propagation is shaped by structural and time-varying context factors.<n>We present an environment-aware, statistically disciplined path loss framework evaluated using leakage-safe cross-validation.
  arXiv  Detail & Related papers  (2025-10-05T20:14:48Z)
• Ensemble Threshold Calibration for Stable Sensitivity Control [0.0]
  We present an end-to-end framework that achieves exact recall with sub-percent variance over tens of millions of geometry pairs.<n>Our approach consistently hits a recall target within a small error, decreases redundant verifications relative to other calibrations, and runs end-to-end on a single TPU v3 core.
  arXiv  Detail & Related papers  (2025-10-02T15:22:28Z)
• Unlocking Dataset Distillation with Diffusion Models [12.306628722616631]
  LD3M learns gradient-based distilled latents and class embeddings end-to-end through a pre-trained latent diffusion model.<n>Across multiple ImageNet subsets at 128x128 and 256x256, LD3M improves downstream accuracy by up to 4.8 percentage points (1 IPC) and 4.2 points (10 IPC) over the prior state-of-the-art.
  arXiv  Detail & Related papers  (2024-03-06T17:41:41Z)
• SCott: Accelerating Diffusion Models with Stochastic Consistency Distillation [74.32186107058382]
  We propose Consistency Distillation (SCott) to enable accelerated text-to-image generation.<n>SCott distills the ordinary differential equation solvers-based sampling process of a pre-trained teacher model into a student.<n>On the MSCOCO-2017 5K dataset with a Stable Diffusion-V1.5 teacher, SCott achieves an FID of 21.9 with 2 sampling steps, surpassing that of the 1-step InstaFlow (23.4) and the 4-step UFOGen (22.1)
  arXiv  Detail & Related papers  (2024-03-03T13:08:32Z)
• SteinDreamer: Variance Reduction for Text-to-3D Score Distillation via Stein Identity [70.32101198891465]
  We show that gradient estimation in score distillation is inherent to high variance. We propose a more general solution to reduce variance for score distillation, termed Stein Score Distillation (SSD) We demonstrate that SteinDreamer achieves faster convergence than existing methods due to more stable gradient updates.
  arXiv  Detail & Related papers  (2023-12-31T23:04:25Z)
• Grad-PU: Arbitrary-Scale Point Cloud Upsampling via Gradient Descent with Learned Distance Functions [77.32043242988738]
  We propose a new framework for accurate point cloud upsampling that supports arbitrary upsampling rates. Our method first interpolates the low-res point cloud according to a given upsampling rate.
  arXiv  Detail & Related papers  (2023-04-24T06:36:35Z)
• Dataset Distillation with Convexified Implicit Gradients [69.16247946639233]
  We show how implicit gradients can be effectively used to compute meta-gradient updates. We further equip the algorithm with a convexified approximation that corresponds to learning on top of a frozen finite-width neural kernel.
  arXiv  Detail & Related papers  (2023-02-13T23:53:16Z)

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.