BossouChimpanzee: Long-term Chimpanzee Video Dataset

Daniel Schofield; Susana Carvalho; Vladimir Iashin; Andrew Zisserman; Max Bain; Arsha Nagrani; David Ng; Claudia Sousa

cs.CVOct 4, 2026

BossouChimpanzee: Long-term Chimpanzee Video Dataset

Authors: Daniel Schofield, Susana Carvalho, Vladimir Iashin, Andrew Zisserman, Max Bain, Arsha Nagrani, David Ng, Claudia Sousa, +5 more

Organizations: Visual Geometry Group, Department of Engineering Science, University of Oxford, UK · University of Oxford, UK · CIBIO–BIOPOLIS, University of Porto, Vairão, Portugal · Department of Science, Gorongosa National Park, Mozambique · Institut de Recherche Environnementale de Guinée (IREG), Bossou, Guinea · University of Rochester, USA · Chubu Gakuin University, Japan · Japan Monkey Centre, Japan · Northwest University, China

Abstract

We describe the BossouChimpanzee video dataset, a unique long-term visual record of wild chimpanzees at an outdoor laboratory for field experiments in Bossou, Guinea, spanning three decades (1988-2018) and comprising over 1,200 hours of continuous video recordings collected through collaborative fieldwork and research. In this paper, we outline the history and scientific contributions of the experimental paradigm and video archive, provide key statistics and details on the structure of the main video dataset, and release an initial ~74h snapshot, BossouChimpanzee70h, covering 23 identified individuals focused on chimpanzee individual and action recognition, ahead of the full video resource. This dataset represents a valuable resource for cognitive and behavioural research in ethology and a rich benchmark for training and evaluating machine learning models on audiovisual data from the wild.

Figures & tables

Figure 1 : A sample of frames from a selection of years of the BossouChimpanzee dataset (1988–2018).

	Face				Body				Action
Year	Boxes	Videos	Tracks	IDs	Boxes	Videos	Tracks	IDs	Videos	Tracks	IDs
2000	1,666,984	46	2,504	21	–	–	–	–	–	–	–
2004	1,436,429	23	1,877	12	–	–	–	–	–	–	–
2006	931,924	20	1,924	11	–	–	–	–	–	–	–
2008	1,607,059	49	2,819	13	–	–	–	–	–	–	–
2012	1,419,441	18	2,973	13	1,878,291	17	2,674	13	11	536	8

Table 1: BossouChimpanzee70h : Year-wise summary. Combined per-year counts across three identity annotation levels. For Face and Body we report the number of videos, tracks, bounding boxes, and individuals (IDs) that appear in a year. The Action columns give videos, tracks, and individuals with additional annotations for spatio-temporal (body-level) action recognition (nut-cracking). The Action Tracks column counts tracks from the earlier body annotation set (Section 3.2 ), not the Body tracks reported in the same row.

Figure 2 : A sample of frames from BossouChimpanzee70h showing Body and Face annotations. Green boxes show Body annotations (SAM 3 tracks [ 35 ] ; Section 3.2 ) and red boxes show Face annotations ( [ 25 , 26 ] ; Section 3.1 ); text labels indicate identity.

	Face				Body				Action
Individual	Boxes	Vids	Tracks	Yrs	Boxes	Vids	Tracks	Yrs	Actions	Vids	Tracks	Yrs
Jeje (M, 1997)	1,154,000	113	1,803	6	384,289	29	355	2	2,287	16	87	2
Jire (F, 1958)	860,987	80	1,056	6	218,196	21	281	2	722	10	66	2
Fanle (F, 1997)	670,403	68	1,201	6	248,469	23	569	2	2,043	13	201	2
Peley (M, 1998)	635,376	81	1,182	5	220,987	11	239	1	1,111	9	33	1
Yolo (M, 1991)	593,501	82	683	4	–	–	–	–	–	–	–	–

Table 2: BossouChimpanzee70h : Individual-wise summary with Face , Body annotation levels, and annotations for spatio-temporal (body-level) Action recognition (nut-cracking). We report the name, sex Male ( M ) or Female ( F ), year of birth, number of annotated bounding boxes, videos, tracks, and years an individual appears in, and the number of action annotations. Body annotations were re-generated using a SAM 3 [ 35 ] + ByteTrack [ 36 ] pipeline; Unknown denotes an on-screen chimpanzee whose identity could not be verified. Actions counts action annotations attributed to a named individual; a further 517 annotations for which the actor could not be identified are not included. The Action Tracks column counts tracks from the earlier body annotation set (Section 3.2 ), not the Body tracks reported in the same row.

	Bouts	Strikes	Failures	Successes	Total
Count	1,281	7,884 (7,367)	588	1,142	10,895

Table 3: BossouChimpanzee70h : Action (nut-cracking) fine-grained annotation counts. The strike count includes 7,367 strikes by identified individuals (in parentheses), and 517 ‘off-screen’ strikes where an actor could not be seen.

Figure 3 : Example of an on-screen nut-cracking action annotation. An on-screen individual, Peley, performs a successful cracking bout comprising 10 strikes. The bout includes one failure when the nut slips away. Another individual, Jeje, passes behind him.

Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

Panel of Table 7 — Table A1 : BossouChimpanzee : Media attribute distributions. Count refers to individual digitised video files, each corresponding to a single archived tape. BossouChimpanzee70h (Table A2 ) instead counts clips excerpted from these recordings, so its per-year count may exceed the archive count while total duration remains a strict subset. Year denotes the field season a recording set is archived under, given by the date of its first video. Seasons spanning a new year (e.g. 2005–2006) are filed under the starting year; Table A2 assigns years by recording date, where such a season appears under the later year.

Panel of Table 8 — Table A2 : BossouChimpanzee70h : Media attribute distributions.

Explore similar work

Automating Visual Recognition of Leprosy in Wild Chimpanzees
Jul 19, 2026Katie I. Murray, Anna C. Bowland, Marina Ramon +9Df-Wild Dataset Roadside Cameras
University of Exeter, Centre for Ecology and Conservation, Penryn, UK · Wild Chimpanzee Foundation, Leipzig, Germany · University of Bristol, School of Computer Science, Bristol, UK
The PanAf-SBR Dataset: Social Behaviour Recognition for Wild Great Apes
Jul 19, 2026Maciej Braszczok, Otto Brookes, Xiaoxuan Ma +6Df-Wild Dataset Human Activity Recognition
University of Bristol, School of Computer Science, Bristol, UK · Wild Chimpanzee Foundation, Leipzig, Germany · Carnegie Mellon University, Robotics Institute, Pennsylvania. USA +4
WildFin: An In-the-Wild Dataset for Fish Behavioral Recognition
Aug 21, 2026Abigail G. Grassick, Jerome Tze-Hou Hsu, Ethan Lin +10Df-Wild Dataset Fish
Cornell University, 14850 Ithaca, NY · University of Colorado Boulder, 80309 Boulder, CO · HHMI Janelia Research Campus, 20147, Ashburn VA

Jul 19, 2026, cs.CVJ/K move · Enter open · S save

Automating Visual Recognition of Leprosy in Wild Chimpanzees

Katie I. Murray, Anna C. Bowland, Marina Ramon, Elena Bersacola+8

University of Exeter, Centre for Ecology and Conservation, Penryn, UK · Wild Chimpanzee Foundation, Leipzig, Germany · University of Bristol, School of Computer Science, Bristol, UK

Df-Wild Dataset Roadside Cameras

Open paper AskPDF arXiv

Leprosy (Mycobacterium leprae) has been confirmed in wild western chimpanzees (Pan troglodytes verus) in West Africa, presenting as clear and progressive visual symptoms. Manual review of camera-trap footage at landscape scale is infeasible, motivating the need for automated screening. We present the first deep learning pipeline for wildlife leprosy detection and contribute the PanLep300 dataset of 125,670 annotated bounding-box crops across 953 tracks from 303 camera-trap videos with ecologically-motivated splits that withhold whole individuals and camera installations. We benchmark spatial (2D), temporally aggregated (2.5D), and video-based (3D) classification approaches to investigate which approach is best suited to automated leprosy detection in wild apes. We find that simple aggregation of crop-level predictions consistently matches or outperforms both learned temporal models and end-to-end video architectures -- consistent with leprosy's static cutaneous presentation. We further find that performance is suppressed when tracklets contain frames of partially visible individuals -- as commonly occurs at the start and end of a track -- and demonstrate that this can be addressed through targeted construction and aggregation strategies.

Abstract

Explore similar work

Automating Visual Recognition of Leprosy in Wild Chimpanzees

The PanAf-SBR Dataset: Social Behaviour Recognition for Wild Great Apes

WildFin: An In-the-Wild Dataset for Fish Behavioral Recognition