Repository navigation
Add shared UResNet for segmentation, PPN, and fragmentation - #190
Open
francois-drielsma wants to merge 4 commits into
Open
francois-drielsma wants to merge 4 commits into
francois-drielsma wants to merge 4 commits into
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #190 +/- ##
==========================================
Coverage 100.00% 100.00%
==========================================
Files 440 449 +9
Lines 39336 40142 +806
==========================================
+ Hits 39336 40142 +806
Flags with carried forward coverage won't be shown. Click here to find out more.
🚀 New features to boost your workflow:
|
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
Add
uresnet_multito run semantic segmentation, point proposal, and Graph-SPICE fragmentation from a single UResNet backbone, with model/loss registration, full-chain integration, documentation, and training/inference configs.Support four semantic coupling modes:
hard: construct separate graphs for predicted semantic classes.independent: construct graphs without semantic conditioning and pool voxel probabilities for final fragment classes.fragment_first: classify fragments using pooled backbone features.soft: condition edge scores on semantic probabilities.Final fragment classes are propagated to member voxels to ensure semantic coherence, while preserving raw voxel predictions. Extract reusable Graph-SPICE heads and unbalanced UResNet+PPN losses, preserving standalone interfaces and checkpoint parameter paths.
Motivation and Context
Enable S3DF comparisons of shared-backbone reconstruction against the existing UResNet+PPN and Graph-SPICE baseline, including the effect of semantic coupling on fragmentation.
Type of Change
How Has This Been Tested?
Focused regression command inside the SPINE Docker image, with the repository on
PYTHONPATH:Model tests ran on CPU using
ghcr.io/deeplearnphysics/spine:latestunder Linux/amd64 emulation, with PyTorch 2.5.1+cu121 and MinkowskiEngine.Checklist
Additional Notes
S3DF training and baseline comparisons remain to be run; no reconstruction-quality or runtime improvement is claimed yet. Configs are under
config/uresnet/multi/. Inputs require upstream deghosting, and pooled/learned fragment semantics require clustering during training and inference.