5 - Estimation de mouvement par maillage actif avec prise en compte de discontinuités
Bibliographic record
Abstract
We aim at modeling a motion vector field by processing a sequence of images. We focus on the detection of motion discontinuities experimented by a moving deformable object. The method is based on a multiscale approach. A Markov Random (MR) label field is built at each scale from an initial distribution of the field. The ground of the motion estimation is a spatial partition of the image given by an elastic mesh superimposed onto the data. The mesh deforms, driven by some selected image features (intensity gradients), under the constraint that the motion field remains locally coherent and uniform within each patch. The motion vector field and the elastic mesh are obtained by minimizing a non-convex energy function considering the image features and the motion vector field simultaneously. Each term of the energy function is defined in a multiscale Markovian context and minimized according to the maximum a posteriori (MAP) criterion. The mesh deformation and the modeling of the related vector field both contribute to the iterative « top-down » optimization process within an alternate relaxation scheme. The model copes with discontinuities thanks to the adaptive partition of the image. The ridges of the mesh progressively move toward the motion discontinuities. The results on noisy complex synthetic sequences show a good estimation of the motion vector field with strong discontinuities at the object interfaces. We apply the proposed method to real short-axis IRM cardiac sequence.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".