Autocalibration: Finding Infinity in a Projective Reconstruction
Bibliographic record
Abstract
In order to extract accurate 3D models from uncalibrated image data it is necessary to upgrade the generated projective reconstructions to a metric space, a process known as auto calibration. The key challenge associated with auto calibration is the nonlinear optimization of a cost function based on extracting camera intrinsics from a potential upgrading transform, and evaluating fitness with respect to prior knowledge of physical cameras. The nonlinearity of the problem leads, in general, to poor convergence and a failure of the calibration process. This paper presents a novel auto calibration pipeline that seeks to develop a more robust approach to the nonlinear optimization. After testing a variety of methods, none of which yielded satisfactory solutions, we have developed a strategy combining the best aspects of two methods representing the current state of the art. The former method preconditions the projective space by ensuring it is quasi-affine with respect to camera centers, allowing a naive initialization in the new space, and uses a fitness measure resistant to focal length collapse. The latter method initializes using the best results of an exhaustive search over reasonable values of focal length. Our novel approach, presented here, uses the exhaustive search initialization of the latter combined with the improved fitness measure of the former, producing results that outperform both of its predecessors.
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 | 0.002 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 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".