Paper
VFM-Recon: Unlocking Cross-Domain Scene-Level Neural Reconstruction with Scale-Aligned Foundation Priors
Authors
Yuhang Ming, Tingkang Xi, Xingrui Yang, Lixin Yang, Yong Peng, Cewu Lu, Wanzeng Kong
Abstract
Scene-level neural volumetric reconstruction from monocular videos remains challenging, especially under severe domain shifts. Although recent advances in vision foundation models (VFMs) provide transferable generalized priors learned from large-scale data, their scaleambiguous predictions are incompatible with the scale consistency required by volumetric fusion. To address this gap, we present VFMRecon, the first attempt to bridge transferable VFM priors with scaleconsistent requirements in scene-level neural reconstruction. Specifically, we first introduce a lightweight scale alignment stage that restores multiview scale coherence. We then integrate pretrained VFM features into the neural volumetric reconstruction pipeline via lightweight task-specific adapters, which are trained for reconstruction while preserving the crossdomain robustness of pretrained representations. We train our model on ScanNet train split and evaluate on both in-distribution ScanNet test split and out-of-distribution TUM RGB-D and Tanks and Temples datasets. The results demonstrate that our model achieves state-of-theart performance across all datasets domains. In particular, on the challenging outdoor Tanks and Temples dataset, our model achieves an F1 score of 70.1 in reconstructed mesh evaluation, substantially outperforming the closest competitor, VGGT, which only attains 51.8.
Metadata
Related papers
Fractal universe and quantum gravity made simple
Fabio Briscese, Gianluca Calcagni • 2026-03-25
POLY-SIM: Polyglot Speaker Identification with Missing Modality Grand Challenge 2026 Evaluation Plan
Marta Moscati, Muhammad Saad Saeed, Marina Zanoni, Mubashir Noman, Rohan Kuma... • 2026-03-25
LensWalk: Agentic Video Understanding by Planning How You See in Videos
Keliang Li, Yansong Li, Hongze Shen, Mengdi Liu, Hong Chang, Shiguang Shan • 2026-03-25
Orientation Reconstruction of Proteins using Coulomb Explosions
Tomas André, Alfredo Bellisario, Nicusor Timneanu, Carl Caleman • 2026-03-25
The role of spatial context and multitask learning in the detection of organic and conventional farming systems based on Sentinel-2 time series
Jan Hemmerling, Marcel Schwieder, Philippe Rufin, Leon-Friedrich Thomas, Mire... • 2026-03-25
Raw Data (Debug)
{
"raw_xml": "<entry>\n <id>http://arxiv.org/abs/2603.12657v1</id>\n <title>VFM-Recon: Unlocking Cross-Domain Scene-Level Neural Reconstruction with Scale-Aligned Foundation Priors</title>\n <updated>2026-03-13T05:00:44Z</updated>\n <link href='https://arxiv.org/abs/2603.12657v1' rel='alternate' type='text/html'/>\n <link href='https://arxiv.org/pdf/2603.12657v1' rel='related' title='pdf' type='application/pdf'/>\n <summary>Scene-level neural volumetric reconstruction from monocular videos remains challenging, especially under severe domain shifts. Although recent advances in vision foundation models (VFMs) provide transferable generalized priors learned from large-scale data, their scaleambiguous predictions are incompatible with the scale consistency required by volumetric fusion. To address this gap, we present VFMRecon, the first attempt to bridge transferable VFM priors with scaleconsistent requirements in scene-level neural reconstruction. Specifically, we first introduce a lightweight scale alignment stage that restores multiview scale coherence. We then integrate pretrained VFM features into the neural volumetric reconstruction pipeline via lightweight task-specific adapters, which are trained for reconstruction while preserving the crossdomain robustness of pretrained representations. We train our model on ScanNet train split and evaluate on both in-distribution ScanNet test split and out-of-distribution TUM RGB-D and Tanks and Temples datasets. The results demonstrate that our model achieves state-of-theart performance across all datasets domains. In particular, on the challenging outdoor Tanks and Temples dataset, our model achieves an F1 score of 70.1 in reconstructed mesh evaluation, substantially outperforming the closest competitor, VGGT, which only attains 51.8.</summary>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n <published>2026-03-13T05:00:44Z</published>\n <arxiv:comment>19 pages, 5 figures, 4 tables</arxiv:comment>\n <arxiv:primary_category term='cs.CV'/>\n <author>\n <name>Yuhang Ming</name>\n </author>\n <author>\n <name>Tingkang Xi</name>\n </author>\n <author>\n <name>Xingrui Yang</name>\n </author>\n <author>\n <name>Lixin Yang</name>\n </author>\n <author>\n <name>Yong Peng</name>\n </author>\n <author>\n <name>Cewu Lu</name>\n </author>\n <author>\n <name>Wanzeng Kong</name>\n </author>\n </entry>"
}