Robust Multi-View Camera Calibration for Wide-Baseline Camera Networks

Real-world camera networks are often characterized by very wide baselines covering a wide range of viewpoints. We describe a method not only calibrating each camera se-quence added to the system automatically, but also taking advantage of multi-view correspondences to make the en-tire calibration framework more robust. Novel camera se-quences can be seamlessly integrated into the system at any time, adding to the robustness of future computations. One of the challenges consists in establishing correspon-dences between cameras. Initializing a bag of features from a calibrated frame, correspondences between cameras are established in a two-step procedure. First, affine invariant features of camera sequences are warped into a common coordinate frame and a coarse matching is obtained be-tween the collected features and the incrementally built and updated bag of features. This allows us to warp images to a common view. Second, scale invariant features are ex-tracted from the warped images. This leads to both more numerous and more accurate correspondences. Finally, the parameters are optimized in a bundle adjustment. Adding the feature descriptors and the optimized 3D positions to the bag of features, we obtain a feature-based scene ab-straction, allowing for the calibration of novel sequences and the correction of drift in single-view calibration track-ing. We demonstrate that our approach can deal with wide baselines. Novel sequences can seamlessly be integrated in the calibration framework. 1. Introduction and Relate