Systems and methods for surface normals sensing with polarization
A method of performing surface profilometry includes receiving one or more polarization raw frames of a printed layer of a physical object undergoing additive manufacturing, the one or more polarization raw frames being captured at different polarizations by one or more polarization cameras, extracting one or more polarization feature maps in one or more polarization representation spaces from the one or more polarization raw frames, obtaining a coarse layer depth map of the printed layer, generating one or more surface-normal images based on the coarse layer depth map and the one or more polarization feature maps, and generating a 3D reconstruction of the printed layer based on the one or more surface-normal images.
1. A method of performing surface profilometry, the method comprising:
receiving one or more polarization raw frames of a printed layer of a physical object undergoing additive manufacturing, the one or more polarization raw frames being captured at different polarizations by one or more polarization cameras;
extracting one or more polarization feature maps in one or more polarization representation spaces from the one or more polarization raw frames;
obtaining a coarse layer depth map of the printed layer;
generating one or more surface-normal images based on the coarse layer depth map and the one or more polarization feature maps; and
generating a 3D reconstruction of the printed layer based on the one or more surface-normal images, wherein the obtaining the coarse layer depth map of the printed layer comprises:
receiving a computer-aided-design (CAD) layer model corresponding to the printed layer of the physical object;
aligning the CAD layer model to the printed layer based on one or more keypoints of the CAD layer model; and
updating the one or more surface-normal images based on the aligned CAD layer model.
2. The method of claim 1 , wherein the one or more polarization feature maps in the one or more polarization representation spaces comprise:
a degree of linear polarization (DOLP) image in a DOLP representation space; and
an angle of linear polarization (AOLP) image in an AOLP representation space.
3. The method of claim 1 , wherein the one or more polarization cameras comprise:
a first polarization camera configured to capture a first partition of a print bed on which the printed layer of the physical object resides; and
a second polarization camera at a distance from the first polarization camera and configured to capture a second partition of the print bed.
4. The method of claim 3 , wherein the obtaining the coarse layer depth map of the printed layer comprises:
constructing the coarse layer depth map based on a parallax shift between polarization raw frames captured by the first and second polarization cameras.
5. The method of claim 1 , wherein the generating the one or more surface-normal images comprises:
determining a pose of the printed layer with respect to each of the one or more polarization cameras;
transforming the coarse layer depth map into one or more camera spaces corresponding to the one or more polarization cameras to generate one or more transformed coarse layer depth maps; and
correcting the one or more surface-normal images based on the one or more transformed coarse layer depth maps.
6. The method of claim 1 , wherein the generating the 3D reconstruction of the printed layer comprises:
integrating surface normals of the one or more surface-normal images over a sample space to determine a shape of a surface of the printed layer.
7. A surface profilometry system comprising:
one or more polarization cameras comprising a polarizing filter, the one or more polarization cameras being configured to capture polarization raw frames at different polarizations; and
a processing system comprising a processor and memory storing instructions that, when executed by the processor, cause the processor to perform:
receiving one or more polarization raw frames of a printed layer of a physical object undergoing additive manufacturing, the one or more polarization raw frames being captured at different polarizations by the one or more polarization cameras;
extracting one or more polarization feature maps in one or more polarization representation spaces from the one or more polarization raw frames;
obtaining a coarse layer depth map of the printed layer;
generating one or more surface-normal images based on the coarse layer depth map and the one or more polarization feature maps; and
generating a 3D reconstruction of the printed layer based on the one or more surface-normal images, wherein the obtaining the coarse layer depth map of the printed layer comprises:
receiving a computer-aided-design (CAD) layer model corresponding to the printed layer of the physical object;
aligning the CAD layer model to the printed layer based on one or more keypoints of the CAD layer model; and
updating the one or more surface-normal images based on the aligned CAD layer model.
8. The surface profilometry system of claim 7 , wherein the one or more polarization feature maps in the one or more polarization representation spaces comprise:
a degree of linear polarization (DOLP) image in a DOLP representation space; and
an angle of linear polarization (AOLP) image in an AOLP representation space.
9. The surface profilometry system of claim 7 , wherein the one or more polarization cameras comprise:
a first polarization camera configured to capture a first partition of a print bed on which the printed layer of the physical object resides; and
a second polarization camera at a distance from the first polarization camera and configured to capture a second partition of the print bed.
10. The surface profilometry system of claim 8 , wherein the obtaining the coarse layer depth map of the printed layer comprises:
constructing the coarse layer depth map based on a parallax shift between polarization raw frames captured by the first and second polarization cameras.
11. The surface profilometry system of claim 7 , wherein the generating the one or more surface-normal images comprises:
determining a pose of the printed layer with respect to each of the one or more polarization cameras;
transforming the coarse layer depth map into one or more camera spaces corresponding to the one or more polarization cameras to generate one or more transformed coarse layer depth maps; and
correcting the one or more surface-normal images based on the one or more transformed coarse layer depth maps.
12. The surface profilometry system of claim 7 , wherein the generating the 3D reconstruction of the printed layer comprises:
integrating surface normals of the one or more surface-normal images over a sample space to determine a shape of a surface of the printed layer.
13. The surface profilometry system of claim 7 , wherein the memory further stores instructions that, when executed by the processor, cause the processor to further perform:
providing the 3D reconstruction of the printed layer to a 3D printing system as a control feedback,
wherein the 3D printing system being configured to additively manufacture the physical object layer by layer.
14. The surface profilometry system of claim 13 , wherein operations of the one or more polarization cameras, the processing system, and the 3D printing system are synchronized via a synchronization signal.