https://mshields.name/tags/camera/ Recent content in camera on Matthew Shields Hugo -- gohugo.io en-GB © 2022 - 2026 Matthew Shields Wed, 03 Jun 2026 00:00:00 +0000 https://mshields.name/blog/2026-06-03-ros-2-camera-calibration/ Wed, 03 Jun 2026 00:00:00 +0000 https://mshields.name/blog/2026-06-03-ros-2-camera-calibration/ Preamble – The Why Underlying many perception tasks is the need to express the pose of one coordinate frame relative to another and to compose such poses along kinematic and sensing chains. This allows us to take transform sensor obeservations from the reference frame of a sensing element. This is useful. A rigid-body pose can be represented by a homogeneous transformation matrix, see here for more on that. \[\begin{equation} \mathbf{T}^{A}_{B} = \begin{bmatrix} \mathbf{R}^{A}_{B} & \mathbf{t}^{A}_{B} \\ \mathbf{0}^{\top} & 1 \end{bmatrix} \in SE(3), \qquad \mathbf{R}^{A}_{B} \in SO(3),\ \ \mathbf{t}^{A}_{B} \in \mathbb{R}^{3}, \label{eq:se3} \end{equation}\] Where \(\mathbf{R}^{A}_{B}\) is a \(3\times3\) rotation matrix belonging to the special orthogonal group \(SO(3)\) and \(\mathbf{t}^{A}_{B}\) is a translation vector.