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

TL;DR

VGG-T$^3$ introduces a scalable 3D reconstruction method that reduces computational complexity from quadratic to linear in the number of images, enabling fast and accurate scene reconstruction and localization.

Contribution

The paper proposes VGG-T$^3$, a novel approach that distills scene geometry into a fixed-size MLP, achieving linear scaling and significant speed improvements over existing methods.

Findings

Reconstructs 1k images in 54 seconds with 11.6x speed-up

Outperforms other linear-time methods in point map accuracy

Demonstrates effective visual localization with unseen images

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.