Journal
ECOLOGICAL INFORMATICS
Volume 70, Issue -, Pages -Publisher
ELSEVIER
DOI: 10.1016/j.ecoinf.2022.101734
Keywords
Animal density; Animal abundance; Camera trapping; Unmarked animal populations; Automated distance estimation; Animal tracking
Categories
Funding
- German Federal Ministry of Education and Research (Bundesministerium fuer Bildung und Forschung (BMBF), Bonn, Gemany
- AMMOD - Automated Multisensor Stations for Monitoring of BioDiversity [FKZ 01LC1903B]
Ask authors/readers for more resources
This paper introduces a fully automatic method called AUDIT for estimating the distance between camera and observed animals accurately. It leverages state-of-the-art depth estimation and alignment techniques, eliminating the need for reference image comparisons or capturing reference image material. Evaluation on an unseen dataset demonstrates that AUDIT achieves high accuracy in distance estimation.
The ongoing biodiversity crisis calls for accurate estimation of animal density and abundance to identify sources of biodiversity decline and effectiveness of conservation interventions. Camera traps together with abundance estimation methods are often employed for this purpose. The necessary distances between camera and observed animals are traditionally derived in a laborious, fully manual or semi-automatic process. Both approaches require reference image material, which is both difficult to acquire and not available for existing datasets. We propose a fully automatic approach we call AUtomated DIstance esTimation (AUDIT) to estimate camera-to-animal distances. We leverage existing state-of-the-art relative monocular depth estimation and combine it with a novel alignment procedure to estimate metric distances. AUDIT is fully automated and requires neither the comparison of observations in camera trap imagery with reference images nor capturing of reference image material at all. AUDIT therefore relieves biologists and ecologists from a significant workload. We evaluate AUDIT on a zoo scenario dataset unseen during training where we achieve a mean absolute distance estimation error over all animal instances of only 0.9864 m and mean relative error (REL) of 0.113. The code and usage instructions are available at https://github.com/PJ-cs/DistanceEstimationTracking
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available