Dataset Generation and Bonobo Classification from Weakly Labelled Videos
This work addresses a domain-specific problem for primatology by enabling automated bonobo identification, but it is incremental as it applies existing methods to a new dataset.
The paper tackled bonobo detection and classification from weakly labeled videos to enable automated testing in enclosures, achieving a best classification accuracy of 75% using a fine-tuned ResNet model after proper data separation.
This paper presents a bonobo detection and classification pipeline built from the commonly used machine learning methods. Such application is motivated by the need to test bonobos in their enclosure using touch screen devices without human assistance. This work introduces a newly acquired dataset based on bonobo recordings generated semi-automatically. The recordings are weakly labelled and fed to a macaque detector in order to spatially detect the individual present in the video. Handcrafted features coupled with different classification algorithms and deep-learning methods using a ResNet architecture are investigated for bonobo identification. Performance is compared in terms of classification accuracy on the splits of the database using different data separation methods. We demonstrate the importance of data preparation and how a wrong data separation can lead to false good results. Finally, after a meaningful separation of the data, the best classification performance is obtained using a fine-tuned ResNet model and reaches 75% of accuracy.