Alexandre-Delplanque/HerdNet

Code for paper "From Crowd to Herd Counting: How to Precisely Detect and Count African Mammals using Aerial Imagery and Deep Learning?" # HerdNet Code for paper "From Crowd to Herd Counting: How to Precisely Detect and Count African Mammals using Aerial Imagery and Deep Learning?" ## Model Architecture ## Detection Examples ## License HerdNet is available under the `MIT License` and is thus open source and freely available. For a complete list of package dependencies with copyright and license info, please look at the file `packages.txt` ## Citation If you use this code in your work, please cite our paper: ``` @article{ title = {From crowd to herd counting: How to precisely detect and count African mammals using aerial imagery and deep learning?}, journal = {ISPRS Journal of Photogrammetry and Remote Sensing}, volume = {197}, pages = {167-180}, year = {2023}, issn = {0924-2716}, doi = {doi.org, url = {sciencedirect.com, author = {Alexandre Delplanque and Samuel Foucher and Jérôme Théau and Elsa Bussière and Cédric Vermeulen and Philippe Lejeune} } ``` ## Pretrained Models Models were trained separatly for each of the two datasets. These pre-trained models follow the (`CC BY-NC-SA-4.0`) license and are available for academic research purposes only, no commercial use is permitted. | Model | Params | Dataset | Environment | Species | F1score | MAE¹ | RMSE² | AC³ | Download | | ------- |:------:| ------------------------------------------------------------ | ---- | ------------------------------------------------ |:-------:|:----:|:-----:|:-----:|:--------------------------------------------------------------------------------------------:| | HerdNet | 18M | Ennedi 2019 | Desert, xeric shrubland and grassland | Camel, donkey, sheep and goat | 73.6% | 6. …