PanAf20K: A Large Video Dataset for Wild Ape Detection and Behaviour Recognition

International Journal of Computer Vision, Год журнала: 2024, Номер 132(8), С. 3086 - 3102

Опубликована: Март 4, 2024

Abstract We present the PanAf20K dataset, largest and most diverse open-access annotated video dataset of great apes in their natural environment. It comprises more than 7 million frames across $$\sim $$ ∼ 20,000 camera trap videos chimpanzees gorillas collected at 18 field sites tropical Africa as part Pan African Programme: The Cultured Chimpanzee. footage is accompanied by a rich set annotations benchmarks making it suitable for training testing variety challenging ecologically important computer vision tasks including ape detection behaviour recognition. Furthering AI analysis information critical given International Union Conservation Nature now lists all species family either Endangered or Critically Endangered. hope can form solid basis engagement community to improve performance, efficiency, result interpretation order support assessments presence, abundance, distribution, thereby aid conservation efforts. code are available from project website:

Язык: Английский

---

Advanced intelligent monitoring technologies for animals: A survey

Neurocomputing, Год журнала: 2024, Номер 585, С. 127640 - 127640

Опубликована: Апрель 4, 2024

Язык: Английский

---

To drum or not to drum: Selectivity in tree buttress drumming by chimpanzees (Pan troglodytes verus) in the Nimba Mountains, Guinea

American Journal of Primatology, Год журнала: 2022, Номер 84(7)

Опубликована: Апрель 5, 2022

Abstract Chimpanzees live in fission‐fusion social organizations, which means that party size, composition, and spatial distribution are constantly flux. Moreover, chimpanzees use a remarkably extensive repertoire of vocal nonvocal forms communication, thought to help convey information such socially spatially dynamic setting. One proposed form communication is buttress drumming, an individual hits tree with its hands and/or feet, thereby producing low‐frequency acoustic signal. It often presumed this behavior functions communicate over long distances is, therefore, goal‐oriented. If so, we would expect exhibit selectivity the choice trees buttresses used drumming. Selectivity key attribute many other goal‐directed chimpanzee behaviors, as nut‐cracking ant dipping. Here, investigate whether at Seringbara study site Nimba Mountains, Guinea, West Africa, show their drumming behavior. Our results indicate more likely larger select thinner have greater surface area. These findings imply not random act, but rather goal‐oriented requires knowledge suitable buttresses. also point long‐distance probable function based on for characteristics impact sound propagation. This provides foundation further assessing cognitive underpinnings wild chimpanzees.

Язык: Английский

---

Highly precise community science annotations of video camera‐trapped fauna in challenging environments

Remote Sensing in Ecology and Conservation, Год журнала: 2024, Номер unknown

Опубликована: Июнь 24, 2024

Abstract As camera trapping grows in popularity and application, some analytical limitations persist including processing time accuracy of data annotation. Typically images are recorded by traps although videos becoming increasingly collected even though they require much more for To overcome with image annotation, trap studies linked to community science (CS) platforms. Here, we extend previous work on CS annotations from a challenging environment; dense tropical forest low visibility high occlusion due thick canopy cover bushy undergrowth at the level. Using platform Chimp&See, established classification 599 956 video clips Africa, assess annotation precision comparing 13 531 1‐min professional ecologist (PE) output 1744 registered, as well unregistered, Chimp&See scientists. We considered 29 categories, 17 species 12 higher‐level which phenotypically similar were grouped. Overall, was 95.4%, increased 98.2% when aggregating groups together. Our findings demonstrate competence scientists working environments hold great promise future animal behaviour, interaction dynamics population monitoring.

Язык: Английский

---

The ontogeny of chimpanzee technological efficiency

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2024, Номер unknown

Опубликована: Июль 31, 2024

Abstract Primate extractive foraging requires years of dedicated learning. Throughout this period, learners peer at conspecifics engaging in the behaviour (“models”), interacting with model and their tools, sometimes stealing freshly extracted resource. This also corresponds to an extended period tolerance from models. Yet long-term effect variation experiences during on technological efficiency individuals is unknown for primate tool use, no research has assessed role both learner model(s) generating individual differences. Using >680 hours video spanning 25 years, we whether stone use social learning (“early period”; ages 0–5) predicted post-early (ages 6+) wild chimpanzees Bossou, Guinea. We found that varied how frequently they peered models’ whole nut-cracking bouts, many opportunities mothers presented, amount intolerance experienced all selected Learners who more became less efficient users, whereas were exposed efficient. Peering bout decreased subsequent efficiency, hinting acquiring cultural components behaviour. Our findings highlight acquisition support view within a tolerant environment are key explaining emergence maintenance complex forms technology. Significance Statement The capacity inclination learn others, along provided by groupmates, thought have enabled evolution technology primates, including hominins. influence remains non-human primates but significant implications transmission evolution. provide longitudinal hypothesis exposure development predicts efficiency. Moreover, show low amounts tolerance, not just general ontogeny Finally, find aspects behavioural relating accurate traits rather than tools efficiently.

Язык: Английский

---

Old age variably impacts chimpanzee engagement and efficiency in stone tool use

Опубликована: Фев. 17, 2025

We know vanishingly little about how long-lived apes experience senescence in the wild, particularly with respect to their foraging behaviors, which are essential for survival. Some wild use tools during and, given additional cognitive and physical challenges presented by tool use, we predict that such behaviors at a heightened risk of senescence. However, until present, longitudinal analysis effects progressive aging on ape tool-use has not been possible due lack available data. In response this research gap, sampled data from video archive contained footage chimpanzees engaging one most complex forms - cracking hard-shelled nuts hammers anvil stones, termed nut an ‘outdoor laboratory’ Bossou, Guinea. By sampling over 17-year period, describe old age influences extent engage – efficiently perform cracking. With increasingly age, began attending experimental sites less frequently than younger individuals. Several elderly exhibited reductions efficiency across multiple components cracking, including taking more time select stone prior longer crack open consume associated pieces kernel. Two also using streamlined behavioral sequences nuts, greater number actions (such as numerous strikes hammer stone). Most notably, report interindividual variability senesced ranging small profound engagement well differences specific aspects each individual. discuss causes these changes reference into captive primates, provide future directions primate both settings.

Язык: Английский

---

Old age variably impacts chimpanzee engagement and efficiency in stone tool use

Опубликована: Фев. 17, 2025

We know vanishingly little about how long-lived apes experience senescence in the wild, particularly with respect to their foraging behaviors, which are essential for survival. Some wild use tools during and, given additional cognitive and physical challenges presented by tool use, we predict that such behaviors at a heightened risk of senescence. However, until present, longitudinal analysis effects progressive aging on ape tool-use has not been possible due lack available data. In response this research gap, sampled data from video archive contained footage chimpanzees engaging one most complex forms - cracking hard-shelled nuts hammers anvil stones, termed nut an ‘outdoor laboratory’ Bossou, Guinea. By sampling over 17-year period, describe old age influences extent engage – efficiently perform cracking. With increasingly age, began attending experimental sites less frequently than younger individuals. Several elderly exhibited reductions efficiency across multiple components cracking, including taking more time select stone prior longer crack open consume associated pieces kernel. Two also using streamlined behavioral sequences nuts, greater number actions (such as numerous strikes hammer stone). Most notably, report interindividual variability senesced ranging small profound engagement well differences specific aspects each individual. discuss causes these changes reference into captive primates, provide future directions primate both settings.

Язык: Английский

---

BEHAVE - facilitating behaviour coding from videos with AI-detected animals

Ecological Informatics, Год журнала: 2025, Номер unknown, С. 103106 - 103106

Опубликована: Март 1, 2025

Язык: Английский

---

A computer vision solution for behavioral recognition in red pandas

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 17, 2025

The survival of the red panda, an endangered arboreal mammal, is challenged by two main factors: habitat loss and health risks that contribute to high morbidity mortality. Abnormal behaviors, such as reduced social locomotor behaviors sleep deprivation, are often signals potential problems. Non-invasive behavioral monitoring using computer vision can provide valuable insights advance research welfare practices. This study presents a dataset 3142 images panda behavior, collected motion-activated camera web crawler technology at Bifengxia Wildlife World. proposes improved lightweight efficient YOLOv8 model for behavior recognition. incorporates adaptive histogram equalization GMBottleNeck module, which enhance detail accentuation reduce parameters. training process was enhanced through integration SimAM attention mechanism feature fusion learning. aforementioned enhancements led YOLOv8s-Red Panda attaining 90.6% accuracy rate, representing 1.4% improvement 1/3 reduction in size comparison data-enhanced baseline (YOLOv8s-DE). exhibits exemplary performance recognition with significantly healthcare optimize animal welfare.

Язык: Английский

---

The MacqD deep-learning-based model for automatic detection of socially housed laboratory macaques

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Апрель 7, 2025

Abstract Despite advancements in video-based behaviour analysis and detection models for various species, existing methods are suboptimal to detect macaques complex laboratory environments. To address this gap, we present MacqD, a modified Mask R-CNN model incorporating SWIN transformer backbone enhanced attention-based feature extraction. MacqD robustly detects their home-cage under challenging scenarios, including occlusions, glass reflections, overexposure light. evaluate compare its performance against pre-existing macaque models, collected analysed video frames from 20 caged rhesus at Newcastle University, UK. Our results demonstrate MacqD’s superiority, achieving median F1-score of 99% with single the focal cage (surpassing next-best by 21%) 90% two macaques. Generalisation tests on different set same animal facility yielded F1-scores 95% 15%) 81% alternative approach 39% ). Finally, was applied videos paired another resulted 90%, reflecting strong generalisation capacity. This study highlights effectiveness accurately detecting across diverse settings.

Язык: Английский

---