Multispecies facial detection for individual identification of wildlife: a case study across ursids
Bibliographic record
Abstract
Abstract To address biodiversity decline in the era of big data, replicable methods of data processing are needed. Automated methods of individual identification (ID) via computer vision are valuable in conservation research and wildlife management. Rapid and systematic methods of image processing and analysis are fundamental to an ever-growing need for effective conservation research and practice. Bears (ursids) are an interesting test system for examining computer vision techniques for wildlife, as they have variable facial morphology, variable presence of individual markings, and are challenging to research and monitor. We leveraged existing imagery of bears living under human care to develop a multispecies bear face detector, a critical part of individual ID pipelines. We compared its performance across species and on a pre-existing wild brown bear Ursus arctos dataset ( BearID ), to examine the robustness of convolutional neural networks trained on animals under human care. Using the multispecies bear face detector and retrained sub-applications of BearID , we prototyped an end-to-end individual ID pipeline for the declining Andean bear Tremarctos ornatus . Our multispecies face detector had an average precision of 0.91–1.00 across all eight bear species, was transferable to images of wild brown bears (AP = 0.93), and correctly identified individual Andean bears in 86% of test images. These preliminary results indicate that a multispecies-trained network can detect faces of a single species sufficiently to achieve high-performance individual classification, which could speed-up the transferability and application of automated individual ID to a wider range of taxa.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".