DFR: Depth from Rotation by Uncalibrated Image Rectification with Latitudinal Motion Assumption
This addresses the challenge of depth estimation from rotating cameras, such as in surveillance, but is incremental as it builds on existing rectification techniques with a new assumption.
The paper tackles the problem of stereo rectification for uncalibrated rotating cameras, which often fails with conventional methods due to rotation-dominant motion and small baselines, and proposes Depth-from-Rotation (DfR), a solution that analytically rectifies images with two-point correspondences, outperforming existing works in effectiveness and efficiency by a significant margin.
Despite the increasing prevalence of rotating-style capture (e.g., surveillance cameras), conventional stereo rectification techniques frequently fail due to the rotation-dominant motion and small baseline between views. In this paper, we tackle the challenge of performing stereo rectification for uncalibrated rotating cameras. To that end, we propose Depth-from-Rotation (DfR), a novel image rectification solution that analytically rectifies two images with two-point correspondences and serves for further depth estimation. Specifically, we model the motion of a rotating camera as the camera rotates on a sphere with fixed latitude. The camera's optical axis lies perpendicular to the sphere's surface. We call this latitudinal motion assumption. Then we derive a 2-point analytical solver from directly computing the rectified transformations on the two images. We also present a self-adaptive strategy to reduce the geometric distortion after rectification. Extensive synthetic and real data experiments demonstrate that the proposed method outperforms existing works in effectiveness and efficiency by a significant margin.