Hi,
Thank you for your wonderful work on OpenSVBRDF! The idea of establishing a mapping between lumitexels and synthesis points is truly creative and inspiring.
I have a question regarding the pipeline in Figure 4:
Are the "measurements" shown in Fig. 4 synthetic (generated from known BRDFs during training), or are they real photometric measurements from physical capture?
If they are real measurements, I understand that the Project operation uses the pre-optimized lighting patterns. Could you kindly clarify:
How exactly is this "pre-optimization" performed? (e.g., objective function, optimization strategy)
Is the code or the final set of 64 lighting patterns available for reproducibility?
Looking forward to your reply!
Best regards,
Jiaxin
Hi,
Thank you for your wonderful work on OpenSVBRDF! The idea of establishing a mapping between lumitexels and synthesis points is truly creative and inspiring.
I have a question regarding the pipeline in Figure 4:
Are the "measurements" shown in Fig. 4 synthetic (generated from known BRDFs during training), or are they real photometric measurements from physical capture?
If they are real measurements, I understand that the Project operation uses the pre-optimized lighting patterns. Could you kindly clarify:
How exactly is this "pre-optimization" performed? (e.g., objective function, optimization strategy)
Is the code or the final set of 64 lighting patterns available for reproducibility?
Looking forward to your reply!
Best regards,
Jiaxin