- ② The third through fifth dimensions of every sample are currently consumed for time and lens samples in pbrt, even though not all scenes need these sample values. For some sample generation algorithms, lower dimensions in the sample are better distributed than higher ones and so this can cause an unnecessary reduction in image quality. Modify pbrt so that the camera can report its sample requirements and then use this information when samples are requested to initialize CameraSamples. Render images and compare results to the current implementation. Do you see an improvement? How do results differ with different samplers? How do you explain any differences you see across samplers?
- ② Keller (2004) and Dammertz and Keller (2008b) described the application of rank-1 lattices to image synthesis. Rank-1 lattices are another way of efficiently generating high-quality low-discrepancy sets of sample points. Read their papers and implement a Sampler based on this approach. Compare results to the other samplers in pbrt.
- ② Implement a Sampler based on orthogonal array sampling, as described by Jarosz et al. (2019). Compare both MSE and Monte Carlo efficiency of this sampler to pbrt’s current samplers.
- ③ Mitchell and Netravali (1988) noted that there is a family of reconstruction filters that use both the value of a function and its derivative at the point to do substantially better reconstruction than if just the value of the function is known. Furthermore, they report that they have derived closed-form expressions for the screen space derivatives of Lambertian and Phong reflection models, although they do not include these expressions in their paper. Investigate derivative-based reconstruction, and extend pbrt to support this technique. If you decide to shy away from deriving expressions for the screen space derivatives for general shapes and BSDF models, you may want to investigate approximations based on finite differencing and the ideas behind the ray differentials of Section 10.1.
- ③ Read some of the papers on adaptive sampling and reconstruction techniques from the “Further Reading” section and implement one of these techniques in pbrt. Note that you will likely need to both write a new Sampler and add additional Film functionality. Measure the effectiveness of the approach you have implemented using Monte Carlo efficiency in order to account for any increased computational cost. How well does your implementation perform compared to non-adaptive samplers?