Search papers, labs, and topics across Lattice.
Queen Mary University of London
3
0
6
Automated identification of individual animals can only be effective if it aligns with ecological questions and data practices, not just algorithmic accuracy.
Training on artificially degraded images boosts wildlife re-identification accuracy by up to 8.5%, even for unseen individuals, offering a simple way to handle noisy real-world data.
Achieve state-of-the-art image reconstruction with a more interpretable and robust approach that is less reliant on training data by learning spatially adaptive sparsity level maps for convolutional dictionaries.
