Fig. 1. Schematic diagram of the perpendicular (s-polarized) and parallel (p-polarized) components of reflected and refracted light

Fig. 2. Schematic diagram of division-of-time polarization imaging system

Fig. 3. Schematic diagram of division-of-amplitude polarization imaging system^[12]

Fig. 4. Schematic diagram of division-of-aperture polarization imaging system^[12]

Fig. 5. Division-of-focal-plane (DoFP) polarization imaging system. (a) Structure schematic diagram; (b) Cooled long-wave infrared polarization imaging thermal imager

Fig. 6. Polarization imaging results of grating metasurface^[33]

Fig. 7. The demosaicking technology based on interpolation. (a) Homogeneity space; (b) The framework of polarization demosaicking based on homogeneity space^[47]

Fig. 8. The demosaicking technology based on NLCSR-CPDM. (a) The flowchart of NLCSR-CPDM; (b) The experimental results on synthetic data; (c) The experimental results on real-captured data^[55]

Fig. 9. The demosaicking technology based on tensor decomposition. (a) The illustration of tensor decomposition for polarization data; (b) Polarization demosaicking model based on sparse tensor factorization^[57]

Fig. 10. The demosaicking technology based on deep learning. (a) The architecture of PDCNN^[58]; (b) The architecture of Fork-Net^[59]; (c) The architecture of CPDNet^[60]; (d) The architecture of TCPDNet^[61]; (e) The architecture of CPDCNN and experimental system^[67]

Fig. 11. The demosaicking technology based on diffusion model. (a) The architecture of DCPM; (b) Results on real-captured data^[68]