Conversation
Collaborator
xiuliren
commented
May 19, 2022
- upgrade python version in ci
* compiles using cmake; we'll use cmake in setup.py later * setup.py works * move CMakeList.txt file to cpp folder * cmake build output file setup * test skeleton fixed * make segmentation a separate module
* output some segmentation with a lot of mergers even with low=0.5 * somehow works, segmentation fault in some run, not stable * watershed is stable without crashing now. memory issue solved.
* use ptrdiff_t to avoid subtraction bug; * fixed the watershed bug by using ptrdiff_t for all the volume size.
* agglomeration kind of works * use size_t for index * greedy mean affinity agglomeration is working
* initial implementation; the fragments got merged, but is not replected in the final segmentation. * minor fix; merge two steps; * works correctly, we only need to use lower min threshold to make the size threshold work * reorganize code and create a folder called segmentation;
* initial implementation; the fragments got merged, but is not replected in the final segmentation. * minor fix; merge two steps; * works correctly, we only need to use lower min threshold to make the size threshold work * reorganize code and create a folder called segmentation; * implemented, runs, but nothing changes. * dilation works, but the segmentation do not look good * fill background with affinity guidance
… then dilate the segmentation, after that we do aggressive agglomeration.
…onversion from pytensor to tensor is not working due to lvalue and rvalue issue
…ut is still not quite good
… neighbors to more neighbors; the segmentation is till not quite good
…erence. The bug might be in the erasion of elements.
…agments with high affinity value
…rative merge code.
…he invalid edge iteration anymore. this looks good, but the result still have many little pieces.
…in machine learning
* merge the small fragments by finding the maximum edge * add voxel number threshold parameter to greedy merge function * we are getting reasonable segmentation
* update test_region_graph * merge array for disjoint sets; also add unit tests for disjoint sets * add to_array function for disjoint sets * use column as segid list, this is more consistent with other formats in machine learning
* add unit test for preprocess; remove object contact is really working * inplace segmentation of watershed * seeded watershed * add robin map github repo as submodule since robin_map is not updated in conda-forge * make the priority queue more general to use * update the edit of priority queue * move priority queue to utils * move priority queue back to segmentation since the functions are built in for the edge type * seeded watershed compiles, but runs with core dump error * compilation in macbook pro * unit test for seeded watershed. the correctness of the result is not checked * add segmentation module with 2d seeded watershed * fixed the bug; the code still has a lot of assertion * remove a lot of assertions * test seeded watershed 3d
* rename greedy_merge to greedy_mean_affinity_agglomeration * get ride of volume boundary flag. we can just use zero-filling and the voxels outside of the volume will have zero segmentation id * get contacting and inner object ids * split the region graph arrays to edge array and sum vector * fix shape assersion bug * merge arrays for region graph
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
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.