Unlocking the Power of Critical Factors for 3D Visual Geometry Estimation

• URL: http://arxiv.org/abs/2604.21713v1
• Date: Thu, 23 Apr 2026 14:20:44 GMT
• Title: Unlocking the Power of Critical Factors for 3D Visual Geometry Estimation
• Abstract summary: This paper investigates the critical factors driving model performance through rigorous ablation studies.<n>We introduce two enhancements to integrate the advantages of optimization-based methods and high-resolution inputs.<n> Experiments on point cloud reconstruction, video depth estimation, and camera pose/intrinsic estimation show that CARVE achieves strong and robust performance across diverse benchmarks.
• Score: 43.14437643346991
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Feed-forward visual geometry estimation has recently made rapid progress. However, an important gap remains: multi-frame models usually produce better cross-frame consistency, yet they often underperform strong per-frame methods on single-frame accuracy. This observation motivates our systematic investigation into the critical factors driving model performance through rigorous ablation studies, which reveals several key insights: 1) Scaling up data diversity and quality unlocks further performance gains even in state-of-the-art visual geometry estimation methods; 2) Commonly adopted confidence-aware loss and gradient-based loss mechanisms may unintentionally hinder performance; 3) Joint supervision through both per-sequence and per-frame alignment improves results, while local region alignment surprisingly degrades performance. Furthermore, we introduce two enhancements to integrate the advantages of optimization-based methods and high-resolution inputs: a consistency loss function that enforces alignment between depth maps, camera parameters, and point maps, and an efficient architectural design that leverages high-resolution information. We integrate these designs into CARVE, a resolution-enhanced model for feed-forward visual geometry estimation. Experiments on point cloud reconstruction, video depth estimation, and camera pose/intrinsic estimation show that CARVE achieves strong and robust performance across diverse benchmarks.

Related papers

• Lightweight Neural Framework for Robust 3D Volume and Surface Estimation from Multi-View Images [45.20850219249498]
  We propose a fully feed-forward framework that regresses scale-normalized volume and surface area directly from multi-view images.<n>By fusing 3D point cloud reconstructions with view-aligned 2D features through a graph-based decoder, our model bypasses iterative optimization.<n>This architecture provides a high-speed, scalable alternative for precise geometric estimation from visual data, maintaining high performance even in resource-constrained or sparse-view scenarios.
  arXiv  Detail & Related papers  (2026-06-22T17:39:18Z)
• Feed-Forward 3D Scene Modeling: A Problem-Driven Perspective [91.23306722968509]
  Generalizable feed-forward 3D reconstruction has witnessed rapid development in recent years.<n>Existing feed-forward approaches share similar high-level architectural patterns.<n>We propose a novel taxonomy centered on model design strategies that are agnostic to the output format.
  arXiv  Detail & Related papers  (2026-04-15T16:07:18Z)
• Towards Geometry-Aware and Motion-Guided Video Human Mesh Recovery [60.51998732898099]
  We introduce HMRMamba, a new paradigm for 3D Human Mesh Recovery.<n>It pioneers the use of Structured State Space Models for their efficiency and long-range modeling prowess.<n>Our framework is distinguished by two core contributions. First, the Geometry-Aware Lifting Module, featuring a novel dual-scan Mamba architecture.
  arXiv  Detail & Related papers  (2026-01-29T08:05:02Z)
• MoRE: 3D Visual Geometry Reconstruction Meets Mixture-of-Experts [50.37005070020306]
  MoRE is a dense 3D visual foundation model based on a Mixture-of-Experts (MoE) architecture.<n>MoRE incorporates a confidence-based depth refinement module that stabilizes and refines geometric estimation.<n>It integrates dense semantic features with globally aligned 3D backbone representations for high-fidelity surface normal prediction.
  arXiv  Detail & Related papers  (2025-10-31T06:54:27Z)
• Gesplat: Robust Pose-Free 3D Reconstruction via Geometry-Guided Gaussian Splatting [21.952325954391508]
  We introduce Gesplat, a 3DGS-based framework that enables robust novel view synthesis and geometrically consistent reconstruction from unposed sparse images.<n>Our approach achieves more robust performance on both forward-facing and large-scale complex datasets compared to other pose-free methods.
  arXiv  Detail & Related papers  (2025-10-11T08:13:46Z)
• Evolving High-Quality Rendering and Reconstruction in a Unified Framework with Contribution-Adaptive Regularization [27.509109317973817]
  3D Gaussian Splatting (3DGS) has garnered significant attention for its high-quality rendering and fast inference speed.<n>Previous methods primarily focus on geometry regularization, with common approaches including primitive-based and dual-model frameworks.<n>We propose CarGS, a unified model leveraging-adaptive regularization to achieve simultaneous, high-quality surface reconstruction.
  arXiv  Detail & Related papers  (2025-03-02T12:51:38Z)
• Poseidon: A ViT-based Architecture for Multi-Frame Pose Estimation with Adaptive Frame Weighting and Multi-Scale Feature Fusion [43.59385149982744]
  Single-frame pose estimation has seen significant progress, but it often fails to capture the temporal dynamics for understanding complex, continuous movements.<n>We propose Poseidon, a novel multi-frame pose estimation architecture that extends the ViTPose model by integrating temporal information.<n>Our approach achieves state-of-the-art performance on the PoseTrack21 and PoseTrack18 datasets, achieving mAP scores of 88.3 and 87.8, respectively.
  arXiv  Detail & Related papers  (2025-01-14T21:34:34Z)
• 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)
• UPose3D: Uncertainty-Aware 3D Human Pose Estimation with Cross-View and Temporal Cues [55.69339788566899]
  UPose3D is a novel approach for multi-view 3D human pose estimation. It improves robustness and flexibility without requiring direct 3D annotations.
  arXiv  Detail & Related papers  (2024-04-23T00:18:00Z)
• Multi-Frame Self-Supervised Depth Estimation with Multi-Scale Feature Fusion in Dynamic Scenes [25.712707161201802]
  Multi-frame methods improve monocular depth estimation over single-frame approaches. Recent methods tend to propose complex architectures for feature matching and dynamic scenes. We show that a simple learning framework, together with designed feature augmentation, leads to superior performance.
  arXiv  Detail & Related papers  (2023-03-26T05:26:30Z)
• Kinematic-Structure-Preserved Representation for Unsupervised 3D Human Pose Estimation [58.72192168935338]
  Generalizability of human pose estimation models developed using supervision on large-scale in-studio datasets remains questionable. We propose a novel kinematic-structure-preserved unsupervised 3D pose estimation framework, which is not restrained by any paired or unpaired weak supervisions. Our proposed model employs three consecutive differentiable transformations named as forward-kinematics, camera-projection and spatial-map transformation.
  arXiv  Detail & Related papers  (2020-06-24T23:56:33Z)

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.