Recognizing European mammals and birds in camera trap images using convolutional neural networks
Abstract
This dataset contains the trained models and the test data set presented in the paper "Recognizing European mammals and birds in camera trap images using convolutional neural networks". In this paper, we present convolutional neural network models based on the EfficientNetV2 and ConvNext architectures to recognize both mammals and bird species in camera trap images.
In the archive files "model_ConvNextBase.tar" and "model_EfficientNetV2.tar", we provide downloads for our best trained models in the Tensorflow2 SavedModel format (https://www.tensorflow.org/guide/saved_model). A script to load and run the models can be found in our Git-Repository: https://github.com/umr-ds/Marburg-Camera-Traps.
We also provide a download for the Marburg Open Forest (MOF) data set consisting of a collection of over 2000 images showing 18 animal species in the archive file "data_MOF.tar". This file contains two folders named "img" and "md", respectively. The "img" folder contains the images grouped in subfolders by recording date and camera trap id. The "md" folder contains the metadata which constists of the bounding box detections for each image. The metadata is grouped into yaml-files for each species.
Metadata
Date | 2023-09-15 |
Authors | Schneider,
Daniel![]() |
Contributors | Lindner,
Kim![]() Vogelbacher, Markus ![]() Bellafkir, Hicham ![]() Mühling, Markus ![]() Farwig, Nina ![]() Freisleben, Bernd ![]() |
Relationship | Is Referenced By:
(URL)
https://github.com/umr-ds/Marburg-Camera-Traps
Is Referenced By: (URL) https://inf-cv.uni-jena.de/wordpress/wp-content/uploads/2023/09/Talk-8-Daniel-Schneider.pdf |
License | Creative Commons Attribution 4.0 |
Faculty | FB12:Department of Mathematics and Computer Science |
Data types | Dataset Model |
URI | https://data.uni-marburg.de/handle/dataumr/246
|
show more | |
show less |
Files
Name | Format | Size | Checksum (MD5) |
---|---|---|---|
data_MOF.tar | .tar | 1.455Gb | 5e9faec564dc3053d8f4999091544733 |
model_ConvNextBase.tar | .tar | 343.5Mb | 3bb8881b72e18d82aab5cec6b5918d43 |
model_EfficientNetV2.tar | .tar | 222.6Mb | 10a57993de643a7b9482260ecb8c7255 |
license_CC-BY-4.0.txt | .txt | 18.21Kb | 380b31767eeb6303e3bc300d8846f180 |
Except where otherwise noted, this item's license is described as Creative Commons Attribution 4.0