SeaCLEF 2016: Object Proposal Classification for Fish Detection in Underwater Videos

This working note describes the results of CVG Jena Fulda team for the fish recognition task in SeaCLEF 2016. Our method is based on convolutional neural networks applied to object proposals for detection as well as species classification. We are using background subtraction proposals that are filtered by a binary SVM classifier for fish detection and a multiclass SVM for species classification. Both SVM’s utilize CNN features extracted from AlexNet. With this pipeline we achieve a recognition precision of 66% and a normalized counting score of 58% on the provided test dataset. We also show that classification of background subtraction proposals works much better for fish detection than background subtraction on its own.