Real-time specularity detection and recovery
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Specularity is a very common phenomenon in the real world and confounds many computer vision tasks such as stereo.The first purpose of this thesis is to design a real-time algorithm of specularity detection.After that, with the knowledge of where the specularities are, a stereo correspondence approach robust to specularity is proposed.Finally, a specularity recovery method is presented to recover the underlying diffuse color using the stereo correspondence information.For real-time specularity detection, a new concept of unnormalized Wiener entropy (UW Entropy) is first proposed in this thesis, which has the desirably simple final form and requires no information about the lighting condition, surface structure, imaging process, pre-segmentation, polarization state, and so forth.However, like other specularity detection methods based on color alone, some false positives may be detected.To distinguish between genuine specularities and false positives, a Support Vector Machine is learned in the proposed SpecLBP space as well as three other spaces as comparisons.An alternative version is also presented for the beam-splitter based stereo pairs in the 3D movie industry, where the curse of side-effect of the beamsplitter is turned into a blessing for identifying problematic specularities.After the genuine specularities are spotted, a new specularity-invariant stereo correspondence method is proposed.By constructing an UW Entropy based matching energy and minimizing it in the MAP-MRF framework using graph cuts, a disparity map robust to specularities can be gained, which offers a precious piece of information for
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 it