Add a benchmark for Passthrough/cropbox vs a Functer filter version

[kunaltyagi]

Passthrough/cropbox vs a Functer filter version

Code is present in the issue, just need to make a PR of that. #4247

Originally posted by @kunaltyagi in #4833 (comment)

[kuspia]

I would like to work on this can u guide me

[mvieth]

@tin1254 You didn't start working on this yet, did you? At least I did not find anything in your pull requests.
@kuspia Have a look at the code in issue #4247 . We would like a benchmark for the PassThrough filter and a benchmark for the CropBox filter (and benchmarks for the functor filter with the same functionality, but that can be a second step). Additionally, have a look at the PassThrough filter tutorial and at the other benchmarks already implemented in the benchmarks folder. Is that a good starting point for you or do you have any specific questions?

[tin1254]

Not yet, I just finished templating the voxel filters benchmark (but i was planning to benchmark the new PassThrough and CropBox soon)

[kuspia]

I understood the concern and need for the Functer filter version now please read what I understood, and if you think I am good to go, say to me for creating a PR so that it gets merged, I have worked on PCL APIs to segment out ROI objects from a noisy 2.5D scan but I am a newbie to GitHub contribution so please suggest if I am going in the wrong direction.

I am supposed to create the following files:
benchmarks/filters/crop_box.cpp
benchmarks/filters/pass_through.cpp

The above files will have an apt piece of code for benchmark and after that, I should make the apt modification in benchmarks/CMakeLists.txt

I have following questions in my mind
A) Now If I am done will this mark this issue closed and my PR will get merged successfully
B) Benchmarks for the functor filter with the same functionality will be solved in this issue or some other

[mvieth]

@kuspia Maybe it is better and easier if you work on another issue. CropBox and PassThrough are currently being rewritten, and we will see if we actually need additional benchmarks for the functor filter. So the goal of this issue is still a bit fuzzy.
Maybe have a look at the good first issues. I am sure there is something interesting there