Let $A$ be an $d \times d$ matrix. The q-numerical range of $A$ is the set: \[ W_q(A) = \left\{ z \in \mathbb{C}: z = \bra{x}A\ket{y}, |\braket{x}{y}| = q, \ket{x} \in \mathbb{C}^d, \ket{y} \in \mathbb{C}^d, \braket{x}{x} = 1, \braket{y}{y} = 1 \right\}. \]

Properties of $W_q(A)$ of a matrix $A$ of dimension $d \times d$ [1],[2]:

1. Note that in the case $q=1$ we get back the standard numerical range;
2. $W_q(A)$ is a convex and bounded set (Tsing theorem) [3];
3. Unitarly invariant: $W_q(A) = W_q(UAU^\dagger)$ for any $U$ unitary matrix;
4. Transpose invariant: $W_q(A) = W_q(A^\top)$;
5. $W_{qz}(A) = zW_q(A)$ for any $z \in \mathbb{C}$, such that $|z|=1$;
6. For any $\mu, \eta \in \mathbb{C}$, we have $W_q(\mu A + \eta \1) = \mu W_q(A) + \eta q$.