Research

Paper

TESTING February 26, 2026

VGG-T$^3$: Offline Feed-Forward 3D Reconstruction at Scale

Authors

Sven Elflein, Ruilong Li, Sérgio Agostinho, Zan Gojcic, Laura Leal-Taixé, Qunjie Zhou, Aljosa Osep

Abstract

We present a scalable 3D reconstruction model that addresses a critical limitation in offline feed-forward methods: their computational and memory requirements grow quadratically w.r.t. the number of input images. Our approach is built on the key insight that this bottleneck stems from the varying-length Key-Value (KV) space representation of scene geometry, which we distill into a fixed-size Multi-Layer Perceptron (MLP) via test-time training. VGG-T$^3$ (Visual Geometry Grounded Test Time Training) scales linearly w.r.t. the number of input views, similar to online models, and reconstructs a $1k$ image collection in just $54$ seconds, achieving a $11.6\times$ speed-up over baselines that rely on softmax attention. Since our method retains global scene aggregation capability, our point map reconstruction error outperforming other linear-time methods by large margins. Finally, we demonstrate visual localization capabilities of our model by querying the scene representation with unseen images.

Metadata

arXiv ID: 2602.23361
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-02-26
Fetched: 2026-02-27 04:35

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.23361v1</id>\n    <title>VGG-T$^3$: Offline Feed-Forward 3D Reconstruction at Scale</title>\n    <updated>2026-02-26T18:59:33Z</updated>\n    <link href='https://arxiv.org/abs/2602.23361v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.23361v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>We present a scalable 3D reconstruction model that addresses a critical limitation in offline feed-forward methods: their computational and memory requirements grow quadratically w.r.t. the number of input images. Our approach is built on the key insight that this bottleneck stems from the varying-length Key-Value (KV) space representation of scene geometry, which we distill into a fixed-size Multi-Layer Perceptron (MLP) via test-time training. VGG-T$^3$ (Visual Geometry Grounded Test Time Training) scales linearly w.r.t. the number of input views, similar to online models, and reconstructs a $1k$ image collection in just $54$ seconds, achieving a $11.6\\times$ speed-up over baselines that rely on softmax attention. Since our method retains global scene aggregation capability, our point map reconstruction error outperforming other linear-time methods by large margins. Finally, we demonstrate visual localization capabilities of our model by querying the scene representation with unseen images.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <published>2026-02-26T18:59:33Z</published>\n    <arxiv:comment>CVPR 2026, Project page: https://research.nvidia.com/labs/dvl/projects/vgg-ttt</arxiv:comment>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Sven Elflein</name>\n    </author>\n    <author>\n      <name>Ruilong Li</name>\n    </author>\n    <author>\n      <name>Sérgio Agostinho</name>\n    </author>\n    <author>\n      <name>Zan Gojcic</name>\n    </author>\n    <author>\n      <name>Laura Leal-Taixé</name>\n    </author>\n    <author>\n      <name>Qunjie Zhou</name>\n    </author>\n    <author>\n      <name>Aljosa Osep</name>\n    </author>\n  </entry>"
}