One View Is Enough! Monocular Training for In-the-Wild Novel View Generation

TL;DR

OVIE is a monocular training method for in-the-wild novel view synthesis that uses unpaired images and a masked training approach, achieving fast inference without explicit 3D models.

Contribution

It introduces a monocular training framework that leverages unpaired images and a masked loss to enable zero-shot novel view synthesis without 3D supervision.

Findings

Outperforms prior methods in zero-shot view synthesis.

Trained on 30 million uncurated images.

Achieves 600x faster inference than previous baselines.

Abstract

Monocular novel-view synthesis has long required multi-view image pairs for supervision, limiting training data scale and diversity. We argue it is not necessary: one view is enough. We present OVIE, trained entirely on unpaired internet images. We leverage a monocular depth estimator as a geometric scaffold at training time: we lift a source image into 3D, apply a sampled camera transformation, and project to obtain a pseudo-target view. To handle disocclusions, we introduce a masked training formulation that restricts geometric, perceptual, and textural losses to valid regions, enabling training on 30 million uncurated images. At inference, OVIE is geometry-free, requiring no depth estimator or 3D representation. Trained exclusively on in-the-wild images, OVIE outperforms prior methods in a zero-shot setting, while being 600x faster than the second-best baseline. Code and models are publicly available at https://github.com/AdrienRR/ovie.