Fin3R: Fine-tuning Feed-forward 3D Reconstruction Models via Monocular Knowledge Distillation

• URL: http://arxiv.org/abs/2511.22429v1
• Date: Thu, 27 Nov 2025 13:10:19 GMT
• Title: Fin3R: Fine-tuning Feed-forward 3D Reconstruction Models via Monocular Knowledge Distillation
• Authors: Weining Ren, Hongjun Wang, Xiao Tan, Kai Han,
• Abstract summary: 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.
• Score: 16.22677555146353
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: We present Fin3R, a simple, effective, and general fine-tuning method for feed-forward 3D reconstruction models. The family of feed-forward reconstruction model regresses pointmap of all input images to a reference frame coordinate system, along with other auxiliary outputs, in a single forward pass. However, we find that current models struggle with fine geometry and robustness due to (\textit{i}) the scarcity of high-fidelity depth and pose supervision and (\textit{ii}) the inherent geometric misalignment from multi-view pointmap regression. Fin3R jointly tackles two issues with an extra lightweight fine-tuning step. We freeze the decoder, which handles view matching, and fine-tune only the image encoder-the component dedicated to feature extraction. The encoder is enriched with fine geometric details distilled from a strong monocular teacher model on large, unlabeled datasets, using a custom, lightweight LoRA adapter. We validate our method on a wide range of models, including DUSt3R, MASt3R, CUT3R, and VGGT. The fine-tuned models consistently deliver sharper boundaries, recover complex structures, and achieve higher geometric accuracy in both single- and multi-view settings, while adding only the tiny LoRA weights, which leave test-time memory and latency virtually unchanged. Project page: \href{http://visual-ai.github.io/fin3r}{https://visual-ai.github.io/fin3r}

Related papers

• TALO: Pushing 3D Vision Foundation Models Towards Globally Consistent Online Reconstruction [57.46712611558817]
  3D vision foundation models have shown strong generalization in reconstructing key 3D attributes from uncalibrated images through a single feed-forward pass.<n>Recent strategies align consecutive predictions by solving global transformation, yet our analysis reveals their fundamental limitations in assumption validity, local alignment scope, and robustness under noisy geometry.<n>We propose a higher-DOF and long-term alignment framework based on Thin Plate Spline, leveraging globally propagated control points to correct spatially varying inconsistencies.
  arXiv  Detail & Related papers  (2025-12-02T02:22:20Z)
• OPFormer: Object Pose Estimation leveraging foundation model with geometric encoding [2.1987601456703474]
  We introduce a unified, end-to-end framework that seamlessly integrates object detection and pose estimation.<n>Our system first employs the CNOS detector to localize target objects.<n>For each detection, our novel pose estimation module, OPFormer, infers the precise 6D pose.
  arXiv  Detail & Related papers  (2025-11-16T14:19:52Z)
• PercHead: Perceptual Head Model for Single-Image 3D Head Reconstruction & Editing [51.56943889042673]
  PercHead is a method for single-image 3D head reconstruction and semantic 3D editing.<n>We develop a unified base model for reconstructing view-consistent 3D heads from a single input image.<n>We highlight the intuitive and powerful 3D editing capabilities of our model through a lightweight, interactive GUI.
  arXiv  Detail & Related papers  (2025-11-04T17:59:15Z)
• Dens3R: A Foundation Model for 3D Geometry Prediction [44.13431776180547]
  Dens3R is a 3D foundation model designed for joint geometric dense prediction.<n>By integrating image-pair matching features with intrinsic invariance modeling, Dens3R accurately regresses multiple geometric quantities.
  arXiv  Detail & Related papers  (2025-07-22T07:22:30Z)
• DiMeR: Disentangled Mesh Reconstruction Model [29.827345186012558]
  DiMeR is a novel geometry-texture disentangled feed-forward model with 3D supervision for sparse-view mesh reconstruction.<n>We streamline the algorithm of mesh extraction by eliminating modules with low performance/cost ratios and redesigning regularization losses with 3D supervision.<n>Extensive experiments demonstrate that DiMeR generalises across sparse-view-, single-image-, and text-to-3D tasks, consistently outperforming baselines.
  arXiv  Detail & Related papers  (2025-04-24T15:39:20Z)
• MUSt3R: Multi-view Network for Stereo 3D Reconstruction [11.61182864709518]
  We propose an extension of DUSt3R from pairs to multiple views, that addresses all aforementioned concerns.<n>We entail the model with a multi-layer memory mechanism which allows to reduce the computational complexity.<n>The framework is designed to perform 3D reconstruction both offline and online, and hence can be seamlessly applied to SfM and visual SLAM scenarios.
  arXiv  Detail & Related papers  (2025-03-03T15:36:07Z)
• 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)
• GeoLRM: Geometry-Aware Large Reconstruction Model for High-Quality 3D Gaussian Generation [65.33726478659304]
  We introduce the Geometry-Aware Large Reconstruction Model (GeoLRM), an approach which can predict high-quality assets with 512k Gaussians and 21 input images in only 11 GB GPU memory. Previous works neglect the inherent sparsity of 3D structure and do not utilize explicit geometric relationships between 3D and 2D images. GeoLRM tackles these issues by incorporating a novel 3D-aware transformer structure that directly processes 3D points and uses deformable cross-attention mechanisms.
  arXiv  Detail & Related papers  (2024-06-21T17:49:31Z)
• Multi-View Large Reconstruction Model via Geometry-Aware Positional Encoding and Attention [54.66152436050373]
  We propose a Multi-view Large Reconstruction Model (M-LRM) to reconstruct high-quality 3D shapes from multi-views in a 3D-aware manner.<n>Specifically, we introduce a multi-view consistent cross-attention scheme to enable M-LRM to accurately query information from the input images.<n>Compared to previous methods, the proposed M-LRM can generate 3D shapes of high fidelity.
  arXiv  Detail & Related papers  (2024-06-11T18:29:13Z)
• 2L3: Lifting Imperfect Generated 2D Images into Accurate 3D [16.66666619143761]
  Multi-view (MV) 3D reconstruction is a promising solution to fuse generated MV images into consistent 3D objects. However, the generated images usually suffer from inconsistent lighting, misaligned geometry, and sparse views, leading to poor reconstruction quality. We present a novel 3D reconstruction framework that leverages intrinsic decomposition guidance, transient-mono prior guidance, and view augmentation to cope with the three issues.
  arXiv  Detail & Related papers  (2024-01-29T02:30:31Z)
• PF-LRM: Pose-Free Large Reconstruction Model for Joint Pose and Shape Prediction [77.89935657608926]
  We propose a Pose-Free Large Reconstruction Model (PF-LRM) for reconstructing a 3D object from a few unposed images. PF-LRM simultaneously estimates the relative camera poses in 1.3 seconds on a single A100 GPU.
  arXiv  Detail & Related papers  (2023-11-20T18:57:55Z)
• Monocular 3D Detection with Geometric Constraints Embedding and Semi-supervised Training [3.8073142980733]
  We propose a novel framework for monocular 3D objects detection using only RGB images, called KM3D-Net. We design a fully convolutional model to predict object keypoints, dimension, and orientation, and then combine these estimations with perspective geometry constraints to compute position attribute.
  arXiv  Detail & Related papers  (2020-09-02T00:51:51Z)

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.