Flow3r: Factored Flow Prediction for Scalable Visual Geometry Learning
This addresses the scalability issue in visual geometry learning for dynamic real-world scenes by reducing reliance on labeled data, though it is an incremental improvement over existing architectures.
The paper tackled the problem of 3D/4D reconstruction requiring expensive dense geometry and pose supervision by introducing Flow3r, a framework that uses dense 2D correspondences (flow) as supervision from unlabeled monocular videos, achieving state-of-the-art results across eight benchmarks with significant gains on in-the-wild dynamic videos.
Current feed-forward 3D/4D reconstruction systems rely on dense geometry and pose supervision -- expensive to obtain at scale and particularly scarce for dynamic real-world scenes. We present Flow3r, a framework that augments visual geometry learning with dense 2D correspondences (`flow') as supervision, enabling scalable training from unlabeled monocular videos. Our key insight is that the flow prediction module should be factored: predicting flow between two images using geometry latents from one and pose latents from the other. This factorization directly guides the learning of both scene geometry and camera motion, and naturally extends to dynamic scenes. In controlled experiments, we show that factored flow prediction outperforms alternative designs and that performance scales consistently with unlabeled data. Integrating factored flow into existing visual geometry architectures and training with ${\sim}800$K unlabeled videos, Flow3r achieves state-of-the-art results across eight benchmarks spanning static and dynamic scenes, with its largest gains on in-the-wild dynamic videos where labeled data is most scarce.