Abstract:Multi-camera systems for structure from motion (SfM) are widely deployed in many mapping applications. Existing solutions assume known rig calibration, synchronized frames among cameras, as well as overlapping field of views (FoVs). In this paper, we derive novel geometric constraints assuming minimal knowns about the multi-camera systems, to benefit low-cost and non-expert use cases where uncalibrated multi-camera systems with non-typical geometry setups present, i.e., no rig calibration, no overlapping FoVs. Assuming that these cameras are co-located and share the same motion of the platform, the proposed constraints utilize the parallelism and length proportionality of motion vectors of these co-located cameras and formulate them as translation constraints into the bundle adjustment (BA). The proposed constraints (called motion constraints) impose a first-order penalty to co-located cameras whose motion speeds and directions between frames do not match. With soft constraints, it can handle loosely synchronized frames (with an error within one second). The proposed constraints are integrated into the BA framework and experimented with different camera setups, i.e., on a group of casually co-located GoPro cameras with no rig calibration, and some with no overlapping views. Our results show that the constraints are extremely effective in improving the reconstruction and pose accuracy for ground motion images: in our self-collected open trajectories without loop closure, the proposed constraints are effective in correcting topographical errors (i.e., trajectory drifts) of the resulting models, and the dense point clouds achieve up to 11.34 m (86.12 %) of mean absolute error (MAE) improvement as compared to reference LiDAR point clouds; our results on KITTI-odometry and KITTI-360 datasets also show an improvement of up to 28.82 m (81.05 %) in terms of the root mean square error (RMSE) of absolute pose error (APE). We expect that the proposed constraints are significant not only as additional geometric constraints for image-based mobile mapping, but also will benefit the broader use of photogrammetry, since it empowers the possibility to harness the traditionally so-called low-quality stereo/multi-camera data (e.g., by non-photogrammetry citizen scientists) into improved 3D products.

Three-Dimensional Reconstruction of A Cylinder Surface Based on Constrained Bundle Adjustment

Multi-view 3D Reconstruction from Uncalibrated Radially-Symmetric Cameras

Vision Based Terrain Reconstruction for Planet Rover Using a Special Binocular Bundle Adjustment

Bundle Adjustment Revisited

Binocular Bundle Adjustment Based Localization and Terrain Stitching for Robot

CoLi-BA: Compact Linearization Based Solver for Bundle Adjustment

Out-of-Core Bundle Adjustment for Large-Scale 3D Reconstruction.

Efficient and Consistent Bundle Adjustment on Lidar Point Clouds

Rba: Reduced Bundle Adjustment for Oblique Aerial Photogrammetry

Efficient Bundle Adjustment for Coplanar Points and Lines.

Local Readjustment for High-Resolution 3D Reconstruction

A Novel Bundle Adjustment Method with Additional Ground Control Point Constraint

Bundle adjustment with motion constraints for uncalibrated multi-camera systems at the ground level

Constrained Bundle Adjustment for Structure From Motion Using Uncalibrated Multi-Camera Systems

PSS-BA: LiDAR Bundle Adjustment with Progressive Spatial Smoothing

$ν$-DBA: Neural Implicit Dense Bundle Adjustment Enables Image-Only Driving Scene Reconstruction

Data-driven Method for 3D Axis-symmetric Object Reconstruction from Single Cone-beam Projection Data

Model-Driven Deep Pipeline With Uncertainty-Aware Bundle Adjustment for Satellite Photogrammetry

On Bundle Adjustment for Multiview PointCloud Registration

A DISTRIBUTED BUNDLE ADJUSTMENT METHOD USING ITERATIVE MOTION AVERAGING