What assumption of identity matrices is this referring to?
motoole2
Good question.
When deriving the essential matrix, we implicitly assumed that the intrinsic matrix $K$ was the identity. As a result, the essential matrix only depends on the extrinsic parameters: a rotation and translation transformation that determines the relative position of the two cameras.
The key difference with the fundamental matrix is that the intrinsic matrix $K$ is generally not an identity matrix, and this needs to be taken into account when working with 2D pixel coordinates.
What assumption of identity matrices is this referring to?
Good question.
When deriving the essential matrix, we implicitly assumed that the intrinsic matrix $K$ was the identity. As a result, the essential matrix only depends on the extrinsic parameters: a rotation and translation transformation that determines the relative position of the two cameras.
The key difference with the fundamental matrix is that the intrinsic matrix $K$ is generally not an identity matrix, and this needs to be taken into account when working with 2D pixel coordinates.