KABR: In-Situ Dataset for Kenyan Animal Behavior Recognition from Drone Videos

Maksim Kholiavchenko, Jenna Kline, Michelle Ramirez, Alec Sheets, Reshma Babu, Namrata Banerji, Matthew Thompson, Nina Van Tiel, Jackson Miliko, Eduardo Bessa, Tanya Berger-Wolf, Daniel Rubenstein, Charles Stewart

Abstract 

We present a novel dataset for animal behavior recognition collected in-situ using video from drones flown over the Mpala Research Centre in Kenya. Videos from D J I Mavic 2S drones flown in January 2023 we reacquired at 5.4K resolution in accordance with IACUC protocols, and processed to detect and track each animal in the frames. An image subregion centered on each animal was extracted and combined in sequence to forma“mini-scene”. Behaviors were then manually labeled for each frame of eachmini-scene by a team of annotators overseen by an expert behavioral ecologist. The resulting labeled mini scenes form our resulting behavior dataset, consisting of more than 10 hours of annotated videos of reticulated giraffes, plains zebras, and Grevy’s zebras, and encompassing seven types of animal behavior and an additional category for occlusions. Benchmark results for state-of-the-art behavioral recognition architectures show labeling accuracy of 61.9% for macro-average (per class),and86.7%for micro-average(per instance).Our dataset complements recent larger, more diverse animal behavior sets and smaller, more specialized ones by being collected in-situ and from drones, both important considerations for the future of animal behavior research. The dataset can be accessed at https://dirtmaxim.github.io/kabr.