Towards Explainable Person Re- Identification
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
Visually recognizing an individual in a crowded area using a distributed camera network is essential for a range of biometric and security applications. We propose a shift in perspective of the ongoing re-identification studies, towards creating more explainable and coherent models that are applicable in real-world engineering problems, even if this comes with a slight decrease in performance. The proposed explainable model uses attribute classification to perform the task of re-identification. This method steps away from intrusive and controversial techniques such as facial recognition to improve public acceptance of re-identification models. Current methods of person re-identification do not explain the importance of each attribute in determining the results, and often use complicated and esoteric algorithms to improve the performance on closed-world datasets which may not represent more realistic open-world scenarios. We applied our approach to the Market-1501 dataset and examined the impact of careful selection of backbone outputs for each individual attribute in our experiment. Our simple model is capable of performing attribute classification for 0-shot re-identification that is explainable and less intrusive when compared to state-of-the-art models focused on re-identification.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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