Accurate automated photo-identification of Near Eastern and European fire salamanders
Authors: Maayan Yesharim, Hava Goldstein, Gabi Kolodny, Assaf Ben-Levy, Tamar Kis-Papo, Dor Gavish, Rotem Vidan, Nirit Lavie-Alon, Olga Rybak, Yoav Ram
Organizations: School of Zoology, Faculty of Life Sciences, Tel Aviv University, Tel Aviv, Israel · Israel Nature and Parks Authority, Israel · The Society for the Protection of Nature in Israel, Tel Aviv, Israel · Safra Center for Bioinformatics, Tel Aviv University, Tel Aviv, Israel
Photo-identification underpins non-invasive capture-recapture monitoring of amphibians, but manual identification becomes slow and error-prone as reference libraries grow. Focusing on the Near Eastern fire salamander (Salamandra infraimmaculata), we developed a pipeline based on zero-shot local-feature matching (SuperPoint and LightGlue) without species-specific training and evaluated it on 2,440 field photographs from Haifa and Tel Dan in northern Israel. We isolated the yellow spot pattern, scored image similarity by the number of high-confidence keypoint matches, which reduced a confound whereby many-spotted individuals accumulate spurious matches, and assigned identities across the database by two-tier clustering. Pipeline-assisted expert review corrected misidentifications, reducing the number of identities by almost a quarter. On held-out individuals, the pipeline ranked the correct individual first for 95.2% of queries and recovered identities with an ARI of 0.958. Without modification, it reached 95.1% and 91.7% top-1 accuracy on two datasets of the European fire salamander (Salamandra salamandra) and recognised 93.7% of recaptures in one of them, compared with 38-77% for five established programs and 93.9% for visual inspection. Zero-shot local-feature matching thus offers amphibian monitoring a re-identification tool that needs no training data, transfers between related species, scales to databases of over a thousand individuals, and can audit existing databases before capture-recapture analysis.
Figures & tables
Dataset
Site
IDs
Images
Singletons
Survey nights
Near Eastern fire salamander
HFA
Wadi Lotem & Haifa Educational Zoo
37
134
11
17
AH
Haifa: Wadi Ahuza
182
376
97
25
NTD 2021–22
Tel Dan Nature Reserve
322
590
179
8
NTD 2022–23
Tel Dan Nature Reserve
553
1,340
298
11
Total
Israel
1,094
2,440
585
61
Table 1: Fire salamander datasets used in this study. Singletons: number of individuals with a single image in the database, excluded as queries in accuracy evaluation. Number of individuals in HFA, AH, and NTD is after identity audit.
Figure 1: Fire salamander dorsal-view photos. Each panel shows 3-4 individuals from one of the datasets of the two species: Near Eastern fire salamander ( S. infraimmaculata ) from Haifa (a) and Tel Dan Nature Reserve (b). European fire salamander ( S. salamandra ) from the Czech Republic (c) and Germany (d). Note the visual difference between species: S. infraimmaculata displays discrete yellow spots, while S. salamandra displays variable markings ranging from discrete spots to elongated stripes.
Figure 2: Image preprocessing applied to two examples. Each row shows one individual. (a, d) Segmentation of body. (b, e) Head orientation: in (a) → (b) only a subtle axis alignment is applied; in (d) → (e) the classifier identifies head-on-right-side and the image is rotated 180∘ . (c, f) Yellow spot extraction. (a,b,c) is individual 349 image 416; (d,e,f) is individual 307 image 228.
Figure 3: Example of the many-spot confound and its correction by confidence filtering. Both panels show the same query (individual 349 from NTD 2021–22, on top) matched against two candidates from the same dataset. Grey lines mark keypoint matches with confidence ∈[0.10,0.65) , which contribute to score0.10 but are excluded by score0.65 ; green lines mark matches with confidence ≥0.65 , counted by score0.65 . (a) Top-1 candidate by score0.10 : a busy false candidate (individual 307, bottom), with score0.10=161 but score0.65=21 ; most keypoint matches are low-confidence. (b) True match from a different survey night, ranked second (top-2) by score0.10=141 but first by score0.65=72 . Using LightGlue’s default confidence threshold c=0.1 incorrectly ranks (a) above (b); score0.65 correctly recovers the true identity.
Evaluation set
Pipeline
Queries
Time (s)
top-1 (%)
top-5 (%)
mAP@R
ARI
Held-out test
Optimized
206
-
95.15
95.63
0.884
0.958
Full dataset
Experimental
1,855
34,509
96.33
98.22
0.868
0.892
Optimized
1,855
35,108
97.47
98.11
0.908
0.958
Table 3: Performance of the optimized pipeline. Held-out test: 206 queries (test images with at least one other image of the same individual in the test set), each matched against the other test images. Full dataset: 2,440 images of 1,094 individuals, including the validation set; 1,855 queries, each matched against all other images from the same site. Time: keypoint extraction and matching on a single GPU, excluding preprocessing. ARI: two-tier clustering; thresholds were fixed on the validation set for the optimized pipeline and calibrated on the experimental pipeline’s own similarity scores for the experimental pipeline.
Figure 4: Same-identity pairs have higher similarity scores than different-identity pairs. Distribution of pairwise similarity scores ( score0.65 ) for the optimized pipeline on the full dataset (2,440 images of 1,094 individuals). Each within-site image pair contributes one score; pairs are never formed across sites. Same-identity pairs (orange, n=3,453 ) and different-identity pairs (grey, n=1,146,843 ) overlap at low scores but not at high scores. Density is plotted on a logarithmic axis. Dashed lines mark the two clustering thresholds, the average-linkage cut ( 29 ) and the must-link threshold ( 55 ), both fixed in advance on the validation set. The must-link threshold admits 77.4% of same-identity pairs while excluding all but 0.0021% of different-identity pairs.
Dataset
top-1 (%)
top-5 (%)
mAP@R
ARI
RR (%)
FRR 1∗ (%)
Czech dataset
95.07
97.09
0.918
0.885
–
–
German dataset
91.74
95.41
0.861
0.768
93.65
2.4
Table 4: Generalization to the European fire salamander datasets. RR and FRR 1∗ are reported only for the German dataset. FRR 1∗ was computed as by Schulte et al. (2024) , dividing rank-1 failures by all images, for comparison with their results; under the common definition ( 1− top-1 accuracy), FRR 1=8.3% .
Similarity score
Definition
AP
AUC
num_matches (raw match count)
∣M(i,j)∣ — all matches, LightGlue’s own confidence floor ( ≥0.10 ), the experimental pipeline’s score
0.882
0.822
mean_confidence
mean of pm over all matches
0.957
0.936
median_confidence
median of pm over all matches
0.954
0.932
p70_confidence
70th-percentile pm
0.956
0.935
p90_confidence
90th-percentile pm
0.955
0.936
mean_top50pct_confidence
mean pm over the more-confident half of matches (those ≥ the pair’s own median)
0.957
0.937
Table S1 : Average precision (AP) and ROC-AUC for 12 candidate similarity scores on the curated pair sample from the validation set (no held-out test images) : all same-identity pairs ( n=3,069 ) balanced against an equal number of hardest negatives (each query’s highest-scoring different-identity candidate under the experimental pipeline). Confidence-filtered counts and fractions are shown at a fixed screening cutoff, c=0.7 . na,nb is the number of SuperPoint keypoints detected in images a,b . AP is used for consistency with every other screening stage in this work; AUC is reported as a secondary reference, since this sample is balanced by construction and so is not itself subject to the class imbalance AP was adopted to address. Confidence-based scores (rows 2–8) are consistently strongest; keypoint-count normalization and geometric verification (rows 9–12), both designed specifically to counter the busy-spot confound, score no better than the raw baseline (AP 0.786 – 0.885 vs. 0.882 ). Boldface marks the three-way tie for best AP (0.957).
All
Same date excluded
Site
nimg
nids
neval
top-1
top-5
mAP@R
neval
top-1
top-5
mAP@R
AH
376
182
279
94.62
94.98
0.865
279
94.62
94.98
0.865
HFA
134
37
123
99.19
99.19
0.961
121
99.17
99.17
0.960
NTD 2021–22
590
322
411
97.81
98.30
0.938
403
97.77
98.26
0.937
NTD 2022–23
1340
553
1042
97.89
98.75
0.902
1034
97.58
98.55
0.901
Overall
2440
1094
1855
97.47
98.11
0.908
1837
97.28
97.99
0.907
Table S2 : Identification accuracy of the optimized pipeline per site on the full dataset. All : the gallery is every other image from the same site. Same date excluded : images of the query’s own individual taken on the same date are also removed from the gallery; images of other individuals from that date remain. neval : query images with at least one other image of the same individual remaining in the gallery.
Figure S1 : Distribution of minimum pairwise image similarity scores across all identities with two or more images ( n=414 ). The solid line shows a KDE of the score distribution, revealing a small cluster of low-scoring identities separated from the main distribution by a local minimum, indicative of possible label errors. The dashed line marks the expert review threshold, set at the local minimum of the KDE ( T=63 ).
AnimalCLEF26 addresses discovery-oriented animal re-identification, where systems must both attach query images to known individuals and discover unseen individuals by clustering them correctly. We present a similarity-to-clustering pipeline for this setting across Eurasian lynx, fire salamander, loggerhead sea turtle, and Texas horned lizard images. The method first isolates the target specimen using segmentation and then applies lightweight species-specific preprocessing for lynx, sea turtle, and salamander images to enhance identity-relevant visual cues, while Texas horned lizard images are used after segmentation only. Pairwise similarities are then estimated with WildFusion by calibrating and combining a MiewID global descriptor with two local matching branches, ALIKED + LightGlue and DISK + LightGlue. The resulting query-query similarities are refined and converted into identity clusters using graph-based clustering, while query-database similarities are used to attach confident samples to known identities. We evaluate training-free and fine-tuned MiewID variants, including Dynamic ArcFace and SphereFace2-Focal adaptations, and combine them in the final ensemble. Our selected ensemble substantially improves on the WildFusion baseline, achieving the best public ARI of 0.72124 and a private ARI of 0.70393, while a simpler preprocessing-before-calibration variant achieves the best private ARI of 0.71087. These results indicate that calibrated global-local fusion with species-aware preprocessing choices is effective for open-set wildlife re-identification under challenging field conditions and visual variation. The implementation code is available on GitHub.
Made In Alexandria Artificial Intelligence Team, Alexandria, Egypt · Faculty of Computers and Data Science, Alexandria University, Alexandria, Egypt · Faculty of Computer Science and Engineering, Alamein International University, New Alamein City, 51718, Egypt +3
Individual animal re-identification from camera-trap imagery is an instance retrieval problem central to non-invasive wildlife monitoring: a query image must retrieve the correct individual from a reference set of known animals. This requires computer vision models to recognize distinctive local patterns in fur, skin, or other visual markings. Current approaches either learn global embeddings as a classification problem, requiring many labeled images per individual while largely ignoring local evidence, or apply off-the-shelf, domain-agnostic image matchers. Although such matchers are pretrained on large and diverse image collections, adapting them to wildlife imagery is challenging because available datasets are small and lack correspondence-level annotations. We study weakly supervised adaptation of a pretrained keypoint matcher using only identity labels, without keypoint-level or geometric correspondence ground truth. We mine informative image pairs with the pretrained matcher, derive weak positive and negative supervision from identity agreement, and contrastively fine-tune the matching network to strengthen correspondences for same-identity pairs and suppress them for different identities. Across open-source wildlife re-identification datasets, our approach improves accuracy over off-the-shelf matchers and a state-of-the-art local--global fusion method. Under an open-world protocol with held-out individuals, it learns a transferable correspondence prior rather than memorizing training identities. To our knowledge, this is the first study of matcher-level, identity-supervised adaptation for animal re-identification. Our method enables data-efficient specialization of image matching models to wildlife domains using identity annotations already available in typical monitoring datasets.
Turhan Can Kargin, Piotr Kubaty, Ekaterina Rostovskaya +3
Faculty of Mathematics and Computer Science, Jagiellonian University, Poland · Doctoral School of Exact and Natural Sciences, Jagiellonian University, Poland · Institute of Environmental Sciences, Faculty of Biology, Jagiellonian University, Kraków, Poland +2
Salmon re-identification in commercial net-pens is challenging due to large populations, which impose strict accuracy requirements and make large-scale labeled data acquisition infeasible. Trajectory IDs can be used as proxy labels, but this introduces trajectory-ID bias. To address these challenges, we propose a patch-based re-identification framework that fuses patch-level predictions into a salmon identity decision. A key component is the prediction of the salmon's lateral line, enabling extraction of texture-anchored patches and patch slices. To enable realistic evaluation, we introduce an experimental setup using multiple cameras placed 6 m apart, allowing the same fish to be recorded in different trajectories. This enables the construction of a cross-camera test set through manual match confirmation. Our ensemble approach outperforms the full-image baseline in same-trajectory validation (0.932 to 0.965 mAP) and cross-camera testing (0.609 to 0.860 mAP). The substantial improvements in the cross-camera setting demonstrate improved generalizability and robustness. Code and data: https://github.com/espenbh/salmon-reid-patch-ensemble.
Espen Uri Høgstedt, Christian Schellewald, Annette Stahl +1
Department of Computer Science, Norwegian University of Science and Technology, Norway · SINTEF Ocean, Norway · Department of Engineering Cybernetics, Norwegian University of Science and Technology, Norway