RidgeSfM: Structure from Motion via Robust Pairwise Matching Under Depth Uncertainty

TL;DR

RidgeSfM introduces a robust method for large-scale indoor scene reconstruction that jointly estimates dense depth maps and camera poses directly from sparse keypoints, outperforming existing SfM pipelines.

Contribution

The paper presents a novel approach that combines deep monocular depth prediction with a modified bundle adjustment to produce dense reconstructions from sparse matches.

Findings

Outperforms state-of-the-art large-scale SfM pipelines.

Capable of aligning hundreds of frames efficiently.

Produces dense depth maps directly from sparse keypoints.

Abstract

We consider the problem of simultaneously estimating a dense depth map and camera pose for a large set of images of an indoor scene. While classical SfM pipelines rely on a two-step approach where cameras are first estimated using a bundle adjustment in order to ground the ensuing multi-view stereo stage, both our poses and dense reconstructions are a direct output of an altered bundle adjuster. To this end, we parametrize each depth map with a linear combination of a limited number of basis "depth-planes" predicted in a monocular fashion by a deep net. Using a set of high-quality sparse keypoint matches, we optimize over the per-frame linear combinations of depth planes and camera poses to form a geometrically consistent cloud of keypoints. Although our bundle adjustment only considers sparse keypoints, the inferred linear coefficients of the basis planes immediately give us dense depth maps. RidgeSfM is able to collectively align hundreds of frames, which is its main advantage over recent memory-heavy deep alternatives that can align at most 10 frames. Quantitative comparisons reveal performance superior to a state-of-the-art large-scale SfM pipeline.