RoMa-$Ω$: What Feed-Forward 3D Models Know About Image Matching

• URL: http://arxiv.org/abs/2609.09507v1
• Date: Tue, 08 Sep 2026 22:50:46 GMT
• Title: RoMa-$Ω$: What Feed-Forward 3D Models Know About Image Matching
• Abstract summary: We show that feed-forward reconstruction models provide strong representations for linear probing and full matching pipelines.<n>We retrain RoMa v2 by replacing its DINO backbone with VGGT-$$.<n>Our resulting model, ours, outperforms state-of-the-art matchers on a wide range of benchmarks.
• Score: 31.653720177504795
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Learned image matching has experienced significant progress in recent years, culminating in robust and accurate matchers such as RoMa, whose robustness is often attributed to its use of frozen DINO features. In a parallel development, feed-forward reconstruction models, such as VGGT, have been trained on ever-growing datasets to accurately regress dense 3D point maps and camera poses. The distinction between matchers and feed-forward reconstruction models has become increasingly blurred with the introduction of matching losses in models such as MASt3R and VGGT-$Ω$. This raises a natural question: what do feed-forward 3D models know about image matching? In this work, we answer this question by analyzing three scenarios: (i) zero-shot matching of patch features, (ii) direct matching of 3D point predictions, and (iii) training a full matcher on top of the learned representations. We find that, despite performing poorly in zero-shot matching, especially in later layers, feed-forward reconstruction models provide strong representations for linear probing and full matching pipelines. We further show that, even without any training, their raw predictions alone enable competitive matching, albeit only under moderate viewpoint changes and modality gaps. Based on these insights, we retrain RoMa v2 by replacing its DINO backbone with VGGT-$Ω$. Our resulting model, \ours, outperforms state-of-the-art matchers on a wide range of benchmarks, e.g. +8.1 mAA compared to RoMa v2 on WxBS.

Related papers

• Free Geometry: Refining 3D Reconstruction from Longer Versions of Itself [38.95274911975506]
  Free Geometry is a framework that enables feed-forward 3D reconstruction models to self-evolve at test time without any 3D ground truth.<n>Our approach consistently improves state-of-the-art foundation models, including Depth Anything 3 and VGGT, across 4 benchmark datasets.
  arXiv  Detail & Related papers  (2026-04-15T16:24:03Z)
• LoMa: Local Feature Matching Revisited [56.73318466794448]
  Local feature matching has long been a fundamental component of 3D vision systems such as Structure-from-Motion (SfM)<n>In this paper, we revisit local feature matching from a data-driven perspective.<n>We combine large and diverse data mixtures, modern training recipes, scaled model capacity, and scaled compute, resulting in remarkable gains in performance.
  arXiv  Detail & Related papers  (2026-04-06T17:59:49Z)
• Fin3R: Fine-tuning Feed-forward 3D Reconstruction Models via Monocular Knowledge Distillation [16.22677555146353]
  Fin3R is a simple, effective, and general fine-tuning method for feed-forward 3D reconstruction models.<n>We validate our method on a wide range of models, including DUSt3R, MASt3R, CUT3R, and VGGT.
  arXiv  Detail & Related papers  (2025-11-27T13:10:19Z)
• RoMa v2: Harder Better Faster Denser Feature Matching [56.71494120301684]
  Dense feature matching aims to estimate all correspondences between two images of a 3D scene.<n>Existing dense matchers fail or perform poorly for many hard real-world scenarios.<n>In this paper, we attack these weaknesses on a wide front through a series of systematic improvements.
  arXiv  Detail & Related papers  (2025-11-19T18:59:38Z)
• 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)
• YoNoSplat: You Only Need One Model for Feedforward 3D Gaussian Splatting [79.38712054342625]
  YoNoSplat is a feedforward model that reconstructs high-quality 3D Gaussian Splatting representations from an arbitrary number of images.<n>Our model is highly versatile, operating effectively with both posed and unposed, calibrated and uncalibrated inputs.<n>It achieves state-of-the-art performance on standard benchmarks in both pose-free and pose-dependent settings.
  arXiv  Detail & Related papers  (2025-11-10T17:21:54Z)
• No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images [100.80376573969045]
  NoPoSplat is a feed-forward model capable of reconstructing 3D scenes parameterized by 3D Gaussians from multi-view images. Our model achieves real-time 3D Gaussian reconstruction during inference. This work makes significant advances in pose-free generalizable 3D reconstruction and demonstrates its applicability to real-world scenarios.
  arXiv  Detail & Related papers  (2024-10-31T17:58:22Z)
• Grounding Image Matching in 3D with MASt3R [8.14650201701567]
  We propose to cast matching as a 3D task with DUSt3R, a powerful 3D reconstruction framework based on Transformers. We propose to augment the DUSt3R network with a new head that outputs dense local features, trained with an additional matching loss. Our approach, coined MASt3R, significantly outperforms the state of the art on multiple matching tasks.
  arXiv  Detail & Related papers  (2024-06-14T06:46:30Z)
• Invariant Training 2D-3D Joint Hard Samples for Few-Shot Point Cloud Recognition [108.07591240357306]
  We tackle the data scarcity challenge in few-shot point cloud recognition of 3D objects by using a joint prediction from a conventional 3D model and a well-trained 2D model. We find out the crux is the less effective training for the ''joint hard samples'', which have high confidence prediction on different wrong labels. Our proposed invariant training strategy, called InvJoint, does not only emphasize the training more on the hard samples, but also seeks the invariance between the conflicting 2D and 3D ambiguous predictions.
  arXiv  Detail & Related papers  (2023-08-18T17:43:12Z)
• RoMa: Robust Dense Feature Matching [17.015362716393216]
  Feature matching is an important computer vision task that involves estimating correspondences between two images of a 3D scene. We propose a model, leveraging frozen pretrained features from the foundation model DINOv2. To further improve robustness, we propose a tailored transformer match decoder.
  arXiv  Detail & Related papers  (2023-05-24T17:59:04Z)

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.