# Nearest-Class Mean and Logits Agreement for Wildlife Open-Set Recognition

arXiv: https://arxiv.org/abs/2510.17338 · HTML: https://mlsanity.com/paper/2510.17338/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 38 references that are in the map.
- First posted on arXiv 20 Oct 2025.

## Builds on (most cited first, 23 of 23)

- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks](https://mlsanity.com/paper/1610.02136/) (2016)
- [Generalized Out-of-Distribution Detection: A Survey](https://mlsanity.com/paper/2110.11334/) (2021)
- [Towards Open Set Deep Networks](https://mlsanity.com/paper/1511.06233/) (2015)
- [Scaling Out-of-Distribution Detection for Real-World Settings](https://mlsanity.com/paper/1911.11132/) (2019)
- [Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation](https://mlsanity.com/paper/2010.16061/) (2020)
- [ViM: Out-Of-Distribution with Virtual-logit Matching](https://mlsanity.com/paper/2203.10807/) (2022)
- [Reducing Network Agnostophobia](https://mlsanity.com/paper/1811.04110/) (2018)
- [BioCLIP: A Vision Foundation Model for the Tree of Life](https://mlsanity.com/paper/2311.18803/) (2023)
- [Adversarial Reciprocal Points Learning for Open Set Recognition](https://mlsanity.com/paper/2103.00953/) (2021)
- [OpenGAN: Open-Set Recognition via Open Data Generation](https://mlsanity.com/paper/2104.02939/) (2021)
- [Efficient Pipeline for Camera Trap Image Review](https://mlsanity.com/paper/1907.06772/) (2019)
- [The Extreme Value Machine](https://mlsanity.com/paper/1506.06112/) (2015)
- [Scaling for Training Time and Post-hoc Out-of-distribution Detection Enhancement](https://mlsanity.com/paper/2310.00227/) (2023)
- [Nearest Neighbor Guidance for Out-of-Distribution Detection](https://mlsanity.com/paper/2309.14888/) (2023)
- [BioCLIP 2: Emergent Properties from Scaling Hierarchical Contrastive Learning](https://mlsanity.com/paper/2505.23883/) (2025)
- [Towards Automatic Wild Animal Monitoring: Identification of Animal Species in Camera-trap Images using Very Deep Convolutional Neural Networks](https://mlsanity.com/paper/1603.06169/) (2016)
- [Wildbook: Crowdsourcing, computer vision, and data science for conservation](https://mlsanity.com/paper/1710.08880/) (2017)
- [WildlifeReID-10k: Wildlife re-identification dataset with 10k individual animals](https://mlsanity.com/paper/2406.09211/) (2024)
- [Calibrated Out-of-Distribution Detection with a Generic Representation](https://mlsanity.com/paper/2303.13148/) (2023)
- [Open-Set Face Recognition with Maximal Entropy and Objectosphere Loss](https://mlsanity.com/paper/2311.00400/) (2023)
- [Improving Wildlife Out-of-Distribution Detection: Africas Big Five](https://mlsanity.com/paper/2506.06719/) (2025)

## Built on by (most cited first, 1 of 1)

- [Multi-Object Tracking Consistently Improves Wildlife Inference](https://mlsanity.com/paper/2605.16672/) (2026)

## Abstract (arXiv)

Current state-of-the-art Wildlife classification models are trained under the closed world setting. When exposed to unknown classes, they remain overconfident in their predictions. Open-set Recognition (OSR) aims to classify known classes while rejecting unknown samples. Several OSR methods have been proposed to model the closed-set distribution by observing the feature, logit, or softmax probability space. A significant drawback of many existing approaches is the requirement to retrain the pre-trained classification model with the OSR-specific strategy. This study contributes a post-processing OSR method that measures the agreement between the models' features and predicted logits. We propose a probability distribution based on an input's distance to its Nearest Class Mean (NCM). The NCM-based distribution is then compared with the softmax probabilities from the logit space to measure agreement between the NCM and the classification head. Our proposed strategy ranks within the top three on two evaluated datasets, showing consistent performance across the two datasets. In contrast, current state-of-the-art methods excel on a single dataset. We achieve an AUROC of 93.41 and 95.35 for African and Swedish animals. The code can be found https://github.com/Applied-Representation-Learning-Lab/OSR.