Puzzles: Unbounded Video-Depth Augmentation for Scalable End-to-End 3D Reconstruction

Jiahao Ma, Lei Wang, Miaomiao Liu, David Ahmedt-Aristizabal, Chuong Nguyen

Keywords: 3D reconstruction, data augmentation

TL;DR: We propose Puzzles, a controllable augmentation that simulates high-quality posed video-depth data from a single image or clip, boosting 3D reconstruction performance with much less training data.

Abstract: Multi-view 3D reconstruction remains a core challenge in computer vision. Recent methods, such as DUSt3R and its successors, directly regress pointmaps from image pairs without relying on known scene geometry or camera parameters. However, the performance of these models is constrained by the diversity and scale of available training data. In this work, we introduce Puzzles, a data augmentation strategy that synthesizes an unbounded volume of high-quality, posed video-depth data from just a single image or video clip. By simulating diverse camera trajectories and realistic scene geometry through targeted image transformations, Puzzles significantly enhances data variety. Extensive experiments show that integrating Puzzles into existing video‑based 3D reconstruction pipelines consistently boosts performance, all without modifying the underlying network architecture. Notably, models trained on only 10% of the original data, augmented with Puzzles, achieve accuracy comparable to those trained on the full dataset.

Primary Area: Deep learning (e.g., architectures, generative models, optimization for deep networks, foundation models, LLMs)

Submission Number: 5767