Automatic object detection for behavioural research using YOLOv8
Uloženo v:
| Název: | Automatic object detection for behavioural research using YOLOv8 |
|---|---|
| Autoři: | Hermens, Frouke |
| Zdroj: | Behav Res Methods Hermens, F 2024, 'Automatic object detection for behavioural research using YOLOv8', Behavior Research Methods, vol. 56, no. 7, pp. 7307-7330. https://doi.org/10.3758/s13428-024-02420-5 |
| Informace o vydavateli: | Springer Science and Business Media LLC, 2024. |
| Rok vydání: | 2024 |
| Témata: | 03 medical and health sciences, 0302 clinical medicine, Surgical tool tracking, 05 social sciences, Video Recording, Humans, YOLO, Original Manuscript, 0501 psychology and cognitive sciences, Automatic object detection, Behavioural analysis, Behavioral Research |
| Popis: | Observational studies of human behaviour often require the annotation of objects in video recordings. Automatic object detection has been facilitated strongly by the development of YOLO (‘you only look once’) and particularly by YOLOv8 from Ultralytics, which is easy to use. The present study examines the conditions required for accurate object detection with YOLOv8. The results show almost perfect object detection even when the model was trained on a small dataset (100 to 350 images). The detector, however, does not extrapolate well to the same object in other backgrounds. By training the detector on images from a variety of backgrounds, excellent object detection can be restored. YOLOv8 could be a game changer for behavioural research that requires object annotation in video recordings. |
| Druh dokumentu: | Article Other literature type |
| Jazyk: | English |
| ISSN: | 1554-3528 |
| DOI: | 10.3758/s13428-024-02420-5 |
| Přístupová URL adresa: | https://pubmed.ncbi.nlm.nih.gov/38750389 https://research.ou.nl/en/publications/4c1a0150-7b06-4045-bdfb-cd0ebec48ad4 https://doi.org/10.3758/s13428-024-02420-5 https://www.scopus.com/pages/publications/85193341452 https://research.ou.nl/en/publications/4c1a0150-7b06-4045-bdfb-cd0ebec48ad4 https://doi.org/10.3758/s13428-024-02420-5 |
| Rights: | CC BY |
| Přístupové číslo: | edsair.doi.dedup.....cc8a479d77ba1ceec497528971c4a860 |
| Databáze: | OpenAIRE |
Full Text Finder 
Nájsť tento článok vo Web of Science | Abstrakt: | Observational studies of human behaviour often require the annotation of objects in video recordings. Automatic object detection has been facilitated strongly by the development of YOLO (‘you only look once’) and particularly by YOLOv8 from Ultralytics, which is easy to use. The present study examines the conditions required for accurate object detection with YOLOv8. The results show almost perfect object detection even when the model was trained on a small dataset (100 to 350 images). The detector, however, does not extrapolate well to the same object in other backgrounds. By training the detector on images from a variety of backgrounds, excellent object detection can be restored. YOLOv8 could be a game changer for behavioural research that requires object annotation in video recordings. |
|---|---|
| ISSN: | 15543528 |
| DOI: | 10.3758/s13428-024-02420-5 |