Table 1 Mapping between real-system calibration matrices and digital twin system parameters
From: Digital-twin-driven unambiguous structured light 3D imaging with physics-aware learning
| Â | Virtual camera | Virtual projector |
|---|---|---|
Intrinsic matrix | ||
Focal length | \({f}_{u}^{c},{f}_{v}^{c}\) | \({f}_{u}^{p},{f}_{v}^{p}\) |
Principle Points | \(({u}_{0}^{c},{v}_{0}^{c})\) | \(({u}_{0}^{p},{v}_{0}^{p})\) |
Skew factor | λc | λp |
Extrinsic matrix | ||
Location | \(-{R}_{c}^{T}{T}_{c}\) | \(-{R}_{p}^{T}{T}_{p}\) |
Rotated angles | \({\psi }^{c}=\arctan \left(\frac{{r}_{32}^{c}}{{r}_{33}^{c}}\right)\) | \({\psi }^{p}=\arctan \left(\frac{{r}_{32}^{p}}{{r}_{33}^{p}}\right)\) |
| Â | \({\theta }^{c}=\arctan \left(\frac{-{r}_{31}^{c}}{\sqrt{{({r}_{32}^{c})}^{2}+{({r}_{33}^{c})}^{2}}}\right)\) | \({\theta }^{p}=\arctan \left(\frac{-{r}_{31}^{p}}{\sqrt{{({r}_{32}^{p})}^{2}+{({r}_{33}^{p})}^{2}}}\right)\) |
| Â | \({\phi }^{c}=\arctan \left(\frac{{r}_{21}^{c}}{{r}_{11}^{c}}\right)\) | \({\phi }^{p}=\arctan \left(\frac{{r}_{21}^{p}}{{r}_{11}^{p}}\right)\) |
