Repository navigation
Conversation
sarlinpe
force-pushed
the
cuts
branch
4 times, most recently
from
October 9, 2026 15:44
f1b84eb to
069d7cf
Compare
sarlinpe
added this pull request to stack #43
October 9, 2026 15:45
…ewers - vidmap/mapper/sub_reconstruction.py: build the 3D-point covisibility graph of the registered keyframes, split it into connected components (edges need at least --min-covisibility-points shared points, components at least --min-model-size images), and extract a standalone pycolmap.Reconstruction per component (cameras, rigs, frames, images, and points with enough observations). - vidmap.run: --split-cuts, --min-covisibility-points (20), --min-model-size (5); export sub-models COLMAP-style (rec/0, rec/1, ...); --html writes embedded viewers when done. - Embedded viewer: write_all_embedded_viewers writes one viewer per sub-model (<name>-sub0.html, ...); larger default point and frustum sizes. - Tests: synthetic two-shot reconstructions (graph, components, extraction, export, viewers).
This branch has not been deployed
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Handle videos with cuts or disconnected sub-sequences by decomposing the final reconstruction into independent sub-models.
vidmap/mapper/sub_reconstruction.py): build the 3D-point covisibility graph of the registered keyframes, keep edges with at least--min-covisibility-pointsshared points, and split it into connected components of at least--min-model-sizeimages. Each component is extracted as a standalonepycolmap.Reconstruction(cameras, rigs, frames, images, and points with enough observations).vidmap.run):--split-cuts,--min-covisibility-points(default 20),--min-model-size(default 5); sub-models are exported COLMAP-style (rec/0,rec/1, ...).--htmlwrites the embedded viewer(s) when mapping finishes.write_all_embedded_viewerswrites one embedded viewer per sub-model (<name>-sub0.html, ...); larger default point and frustum sizes.tests/test_sub_reconstruction.py): synthetic two-shot reconstructions covering graph construction, components, extraction, export, and viewers.Stack: rebased onto the two fixes below it; dropped the unused temporal-window cut detector and the redundant
reconstruct()split arguments (they come fromRunOptions).