Reduce time taken in CropHull by using O(n) pre-filtering with CropBox#3883
Open
DaniilSNikulin wants to merge 24 commits intoPointCloudLibrary:masterfrom
Open
Reduce time taken in CropHull by using O(n) pre-filtering with CropBox#3883DaniilSNikulin wants to merge 24 commits intoPointCloudLibrary:masterfrom
CropHull by using O(n) pre-filtering with CropBox#3883DaniilSNikulin wants to merge 24 commits intoPointCloudLibrary:masterfrom
Conversation
added 4 commits
April 7, 2020 18:37
…udLibrary#1657) * funciton applyFilter didn't clear output data before start adding new point or indices * issue PointCloudLibrary#1657: CropHull 3d not cropping inside. [decision](piaggiofastforward@b8a8a2b)
kunaltyagi
reviewed
Apr 7, 2020
Morwenn
reviewed
Apr 8, 2020
suggestions `resize(0)` -> `clear` from @kunaltyagi Co-Authored-By: Kunal Tyagi <tyagi.kunal@live.com>
kunaltyagi
reviewed
Apr 8, 2020
kunaltyagi
added
changelog: enhancement
kunaltyagi
changed the title
CropHull by using O(n) pre-filtering with CropBox
added 2 commits
April 9, 2020 19:59
… expect extract_removed_indices==true
* delete unused coments of code (2 lines) * delete applyFilter[23]D(PointCloud) -- private and not used * delete applyFilter(PointCloud) -- FilterIndices already have got * implementation * deleted applyFilter fix bug: FilterIndices funtionality for * keepOrganized cloud was missed in original implementation * set extract_removed_indices default to true and support * removed_indices functionality * perfomance testing: PCL.ConvexHull_2dsquare w/o CrobBox spent 76ms, * w/ spent 28ms; PCL.ConvexHull_3dcube w/o CropBox spent 1034ms, w/ * spent 640ms
kunaltyagi
reviewed
Apr 10, 2020
Morwenn
reviewed
Apr 10, 2020
Member
|
Just a reminder to remember to try to keep these PRs as atomic as possible. Avoid the urge to fix the internals of the class unless the PR is all about that. As an example: I would split this PR with its original intent, 1 PR for fixing the original bug, another to change CropHull to prefilter with CropBox. Just keep those things in mind before venturing into optimizing the internals. |
bug was founded by @Morwenn also tests was modified to detect this bug
Co-Authored-By: Kunal Tyagi <tyagi.kunal@live.com>
…keelInside = crop_outside_ ^ negative_
…thub.com/DaniilSNikulin/pcl into issue_3875_using_CropBox_inside_CropHull
SergioRAgostinho
added
needs: code review
kunaltyagi
requested changes
Apr 18, 2020
kunaltyagi
requested changes
Apr 20, 2020
Member
kunaltyagi
left a comment
kunaltyagi
left a comment
There was a problem hiding this comment.
The tests can be rewritten to be more maintainable.
Morwenn
reviewed
Apr 22, 2020
Morwenn
reviewed
Apr 22, 2020
kunaltyagi
reviewed
Apr 22, 2020
test/filters/test_crop_hull.cpp
Outdated
|
|
||
| cropHullFilter.setHullIndices(hullPolygons); | ||
| cropHullFilter.setHullCloud(hullPoints); | ||
| //cropHullFilter.setDim(2); // if you uncomment this, it will work |
kunaltyagi
reviewed
Apr 24, 2020
| pcl::Indices filteredIndices; | ||
| cropHullFilter.filter(filteredIndices); | ||
| pcl::test::EXPECT_EQ_VECTORS(testData.insideIndices, filteredIndices); | ||
| cropHullFilter.filter(filteredIndices); |
Member
There was a problem hiding this comment.
Seems to be duplicated check wrt 2 lines before
Contributor
Author
There was a problem hiding this comment.
this test check reentrancy of function filter. results should not depend on amount of calls before.
PS: maybe I'm not correct use termin of reentrancy and it must rename to repeatability
kunaltyagi
reviewed
Apr 24, 2020
kunaltyagi
reviewed
Apr 24, 2020
| pcl::Indices filteredIndices; | ||
| cropHullFilter.filter(filteredIndices); | ||
| pcl::test::EXPECT_EQ_VECTORS(testData.insideIndices, filteredIndices); | ||
| cropHullFilter.filter(filteredIndices); |
| for (std::size_t index = 0; index < indices_->size (); index++) | ||
| const bool keepInside = crop_outside_ ^ negative_; | ||
|
|
||
| pcl::Indices cropInlierIndices; |
Member
There was a problem hiding this comment.
Let's add a note here why we're doing this since this is the core contribution of the PR.
kunaltyagi
added
needs: pr merge
This was referenced Jul 14, 2021
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Firstly reason for PR was only issue #3875: add CropBox inside CropHull for increasing performance.
In progress was founded and fixed some bugs:
point or indices.
Added unit tests for CropHull. (only ConvexHull case).
Also was deleted some duplicated code. Implementation for cloud and indices filtering was equals, but duplicates.
resolve #1657
resolve #3541
resolve #3960
UPD:
CropHull::applyFilter[23]D(PointCloud)-- private and not usedCropHull::applyFilter(PointCloud)--FilterIndices::applyFilter(PointCloud)already have got implementationCropHull::applyFilterfix bug:FilterIndicesfuntionality forkeepOrganizedcloud was missed in original implementation