Skip to content

add feature registration and slice-to-slice correlation#42

Merged
huangchieh merged 1 commit into
mainfrom
issue40
Jun 20, 2026
Merged

add feature registration and slice-to-slice correlation#42
huangchieh merged 1 commit into
mainfrom
issue40

Conversation

@huangchieh

Copy link
Copy Markdown
Collaborator

No description provided.

Copilot AI review requested due to automatic review settings June 20, 2026 19:03
@huangchieh huangchieh merged commit 4a06742 into main Jun 20, 2026
1 check passed

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

This PR adds dataset-wide analysis and visualization tooling to evaluate style-transfer outputs, including feature-based similarity (Dice) and slice-to-slice correlation, plus scripts to generate combined “sanity check” figures and example output artifacts.

Changes:

  • Add dataWideStyletransAnalysis.py to compute SSIM, feature Dice (including blob-based feature detection), and adjacent-slice correlation metrics and export CSV + summary plots.
  • Add makeCombinedSanityFigures.py to render combined multi-panel sanity figures from the exported CSV metrics and input tar slices (including feature overlays).
  • Add example result artifacts under results/dataWideStyletransAnalysis/ (CSV summary + SVG figure).

Reviewed changes

Copilot reviewed 3 out of 20 changed files in this pull request and generated 2 comments.

File Description
src/preEvaluation/makeCombinedSanityFigures.py New CLI/script to build combined image + metric trend sanity figures (standard, gradient, and feature modes).
src/preEvaluation/dataWideStyletransAnalysis.py New dataset-wide metric computation pipeline producing SSIM/feature/correlation CSVs and summary plots.
results/dataWideStyletransAnalysis/summary_stats.csv Example summary metric export.
results/dataWideStyletransAnalysis/L20_L1/content_preservation.svg Example generated figure export.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

ax_corr.set_xlabel("Adjacent pair index (i vs i+1)")
ax_corr.set_ylabel("Correlation")
ax_corr.grid(alpha=0.25)
ax_corr.legend(loc="bottom right", frameon=False)
Comment on lines +694 to +703
for i in range(len(valid_slice_ids) - 1):
corr_o = compute_flat_correlation(o_imgs[i], o_imgs[i + 1])
corr_t = compute_flat_correlation(t_imgs[i], t_imgs[i + 1])
corr_rows.append(
{
"tar_file": tar_name,
"sample_id": sample_id,
"pair_idx": i,
"slice_idx_i": valid_slice_ids[i],
"slice_idx_ip1": valid_slice_ids[i + 1],
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants