# Towards Zero-Shot Camera Trap Image Categorization

arXiv: https://arxiv.org/abs/2410.12769 · HTML: https://mlsanity.com/paper/2410.12769/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 42 references that are in the map.
- First posted on arXiv 16 Oct 2024.

## Builds on (most cited first, 22 of 22)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Aggregated Residual Transformations for Deep Neural Networks](https://mlsanity.com/paper/1611.05431/) (2016)
- [Billion-scale similarity search with GPUs](https://mlsanity.com/paper/1702.08734/) (2017)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [RandAugment: Practical automated data augmentation with a reduced search space](https://mlsanity.com/paper/1909.13719/) (2019)
- [Swin Transformer V2: Scaling Up Capacity and Resolution](https://mlsanity.com/paper/2111.09883/) (2021)
- [Vision Transformers Need Registers](https://mlsanity.com/paper/2309.16588/) (2023)
- [Recognition in Terra Incognita](https://mlsanity.com/paper/1807.04975/) (2018)
- [The Faiss library](https://mlsanity.com/paper/2401.08281/) (2024)
- [BEiT v2: Masked Image Modeling with Vector-Quantized Visual Tokenizers](https://mlsanity.com/paper/2208.06366/) (2022)
- [BioCLIP: A Vision Foundation Model for the Tree of Life](https://mlsanity.com/paper/2311.18803/) (2023)
- [A deep active learning system for species identification and counting in camera trap images](https://mlsanity.com/paper/1910.09716/) (2019)
- [Perspectives on individual animal identification from biology and computer vision](https://mlsanity.com/paper/2103.00560/) (2021)
- [Danish Fungi 2020 -- Not Just Another Image Recognition Dataset](https://mlsanity.com/paper/2103.10107/) (2021)
- [Pytorch-Wildlife: A Collaborative Deep Learning Framework for Conservation](https://mlsanity.com/paper/2405.12930/) (2024)
- [Bag of Tricks for Long-Tail Visual Recognition of Animal Species in Camera-Trap Images](https://mlsanity.com/paper/2206.12458/) (2022)
- [Mastering Large Scale Multi-label Image Recognition with high efficiency overCamera trap images](https://mlsanity.com/paper/2008.07828/) (2020)

## Built on by (most cited first, 3 of 3)

- [Vision Transformers for Zero-Shot Clustering of Animal Images: A Comparative Benchmarking Study](https://mlsanity.com/paper/2602.03894/) (2026)
- [GreenCrossingAI: A Camera Trap/Computer Vision Pipeline for Environmental Science Research Groups](https://mlsanity.com/paper/2507.09410/) (2025)
- [Bridging Domain Gaps for Fine-Grained Moth Classification Through Expert-Informed Adaptation and Foundation Model Priors](https://mlsanity.com/paper/2508.20089/) (2025)

## Abstract (arXiv)

This paper describes the search for an alternative approach to the automatic categorization of camera trap images. First, we benchmark state-of-the-art classifiers using a single model for all images. Next, we evaluate methods combining MegaDetector with one or more classifiers and Segment Anything to assess their impact on reducing location-specific overfitting. Last, we propose and test two approaches using large language and foundational models, such as DINOv2, BioCLIP, BLIP, and ChatGPT, in a zero-shot scenario. Evaluation carried out on two publicly available datasets (WCT from New Zealand, CCT20 from the Southwestern US) and a private dataset (CEF from Central Europe) revealed that combining MegaDetector with two separate classifiers achieves the highest accuracy. This approach reduced the relative error of a single BEiTV2 classifier by approximately 42% on CCT20, 48% on CEF, and 75% on WCT. Besides, as the background is removed, the error in terms of accuracy in new locations is reduced to half. The proposed zero-shot pipeline based on DINOv2 and FAISS achieved competitive results (1.0% and 4.7% smaller on CCT20, and CEF, respectively), which highlights the potential of zero-shot approaches for camera trap image categorization.