Flow-Corrected Shape Optimization: Taming Manifold Drift in High-Dimensional 3D Models

• URL: http://arxiv.org/abs/2608.07199v1
• Date: Fri, 07 Aug 2026 13:14:35 GMT
• Title: Flow-Corrected Shape Optimization: Taming Manifold Drift in High-Dimensional 3D Models
• Abstract summary: manifold drift is the tendency of gradient-based optimization to move latent vectors away from the manifold of valid shapes.<n>Existing mitigation strategies, including latent regularization and flow-matching approaches, demand a difficult trade-off between objective guidance and generative fidelity.<n>We introduce a novel-corrector framework that alternates between gradient steps for objective 3D shape models to drive the latent state back to the valid shape manifold.
• Score: 30.44524553376579
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Optimizing 3D shapes within the latent spaces of deep generative models is fundamental to computer assisted engineering, yet remains prone to a critical failure mode we term manifold drift: the tendency of gradient-based optimization to move latent vectors away from the manifold of valid shapes. This problem is exacerbated in state-of-the-art 3D shape generative models that operate in increasingly high-dimensional latent spaces where valid shapes occupy a vanishingly small fraction of the full space. Existing mitigation strategies, including latent regularization and flow-matching approaches, either sacrifice expressiveness, demand a difficult trade-off between objective guidance and generative fidelity that remains prone to manifold drift, or are computationally infeasible to scale to modern, large-capacity 3D shape models. We introduce a novel optimizer-corrector framework that alternates between gradient steps for objective minimization and guided flow matching to drive the latent state back to the valid shape manifold. By decoupling objective minimization from flow-based correction, optimizing freely and correcting strictly, this alternating design avoids inherent trade-offs, preserving geometric validity without sacrificing expressiveness while remaining computationally feasible on modern 3D shape models. We demonstrate its effectiveness across generative priors of varying complexity, from simple vector latent spaces to large-scale architectures across a variety of downstream optimization tasks, including aerodynamic drag reduction and object compliance optimization.

Related papers

• Optimization and Generation in Aerodynamics Inverse Design [41.756961128464816]
  Inverse design with physics-based objectives is challenging because it couples high-dimensional geometry with expensive simulations.<n>We revisit inverse design through two canonical solutions, the optimal design point and the optimal design distribution.<n>We propose a new training loss for cost predictors and a density-gradient optimization method that improves objectives while preserving plausible shapes.
  arXiv  Detail & Related papers  (2026-02-03T14:32:26Z)
• Tail-Aware Post-Training Quantization for 3D Geometry Models [58.79500829118265]
  Post-Training Quantization (PTQ) enables efficient inference without retraining.<n>PTQ fails to transfer effectively to 3D models due to intricate feature distributions and prohibitive calibration overhead.<n>We propose TAPTQ, a Tail-Aware Post-Training Quantization pipeline for 3D geometric learning.
  arXiv  Detail & Related papers  (2026-02-02T07:21:15Z)
• 3DID: Direct 3D Inverse Design for Aerodynamics with Physics-Aware Optimization [53.40777496077015]
  Inverse design aims to design the input variables of a physical system to optimize a specified objective function.<n>In 3D domains, the design space grows exponentially, rendering exhaustive grid-based searches infeasible.<n>We propose a 3D Inverse Design framework that directly navigates the 3D design space by coupling a continuous latent representation with a physics-aware optimization strategy.
  arXiv  Detail & Related papers  (2025-12-06T13:09:03Z)
• GLVD: Guided Learned Vertex Descent [18.278508401516426]
  We introduce GLVD, a hybrid method for 3D face reconstruction from few-shot images.<n>It integrates per-vertex neural field optimization with global structural guidance from dynamically predicted 3D keypoints.<n> GLVD achieves state-of-the-art performance in single-view settings and remains highly competitive in multi-view scenarios.
  arXiv  Detail & Related papers  (2025-10-07T15:40:10Z)
• TripOptimizer: Generative 3D Shape Optimization and Drag Prediction using Triplane VAE Networks [4.4288915456583755]
  Tripr is a framework for rapid aerodynamic analysis and shape optimization directly from vehicle point cloud data.<n>Tripr employs a Variational Autoencoder featuring a triplane-based implicit neural representation for high-fidelity 3D geometry reconstruction and a drag coefficient prediction head.
  arXiv  Detail & Related papers  (2025-09-05T15:14:19Z)
• Adjoint-Based Aerodynamic Shape Optimization with a Manifold Constraint Learned by Diffusion Models [12.019764781438603]
  We introduce an adjoint-based aerodynamic shape optimization framework that integrates a diffusion model trained on existing designs to learn a smooth manifold of aerodynamically viable shapes.<n>We demonstrate how AI generated priors integrates effectively with adjoint methods to enable robust, high-fidelity aerodynamic shape optimization through automatic differentiation.
  arXiv  Detail & Related papers  (2025-07-31T11:21:20Z)
• Efficient Design of Compliant Mechanisms Using Multi-Objective Optimization [50.24983453990065]
  We address the synthesis of a compliant cross-hinge mechanism capable of large angular strokes.<n>We formulate a multi-objective optimization problem based on kinetostatic performance measures.
  arXiv  Detail & Related papers  (2025-04-23T06:29:10Z)
• SparseFlex: High-Resolution and Arbitrary-Topology 3D Shape Modeling [79.56581753856452]
  SparseFlex is a novel sparse-structured isosurface representation that enables differentiable mesh reconstruction at resolutions up to $10243$ directly from rendering losses.<n>By enabling high-resolution, differentiable mesh reconstruction and generation with rendering losses, SparseFlex significantly advances the state-of-the-art in 3D shape representation and modeling.
  arXiv  Detail & Related papers  (2025-03-27T17:46:42Z)
• Non-rigid Structure-from-Motion: Temporally-smooth Procrustean Alignment and Spatially-variant Deformation Modeling [34.606331252248886]
  Non-rigid Structure-from-Motion (NRSfM) has been extensively studied and great progress has been made.<n>There are still key challenges that hinder their broad real-world applications.<n>This paper proposes to resolve the above issues from a spatial-temporal modeling perspective.
  arXiv  Detail & Related papers  (2024-05-07T13:33:50Z)
• Learning Versatile 3D Shape Generation with Improved AR Models [91.87115744375052]
  Auto-regressive (AR) models have achieved impressive results in 2D image generation by modeling joint distributions in the grid space. We propose the Improved Auto-regressive Model (ImAM) for 3D shape generation, which applies discrete representation learning based on a latent vector instead of volumetric grids.
  arXiv  Detail & Related papers  (2023-03-26T12:03:18Z)
• Learned Vertex Descent: A New Direction for 3D Human Model Fitting [64.04726230507258]
  We propose a novel optimization-based paradigm for 3D human model fitting on images and scans. Our approach is able to capture the underlying body of clothed people with very different body shapes, achieving a significant improvement compared to state-of-the-art. LVD is also applicable to 3D model fitting of humans and hands, for which we show a significant improvement to the SOTA with a much simpler and faster method.
  arXiv  Detail & Related papers  (2022-05-12T17:55:51Z)

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.