173 Downloads (Pure)

Abstract

We present the first work where re-identification ofthe Giant Sunfish (Mola alexandrini) is automated using computer vision and deep learning. We propose a pipeline that scores an mAP of 60.34% on a full rank of the novel TinyMola dataset which includes 31 IDs and 91 images. The method requires no domain-adaptation or training which makes it especially suited for low-budget or volunteer-based projects, like Match My Mola, as part of a human-in-the-loop model.

The pipeline includes segmentation, keypoint detection and description, keypoint matching, and ranking. The choice of feature descriptor has the largest impact on the performance and we show that the deep learning based SuperPoint descriptor greatly outperforms handcrafted descriptors like SIFT and RootSIFT independent of the segmentation level and matching method. Combining SuperPoint and the graph neural network based SuperGlue matching method produces the best results.
Original languageEnglish
JournalProceedings of the Northern Lights Deep Learning Workshop
Volume3
Number of pages9
ISSN2703-6928
DOIs
Publication statusPublished - 28 Mar 2022
EventNorthern Lights Deep Learning Workshop - Tromsø, Norway
Duration: 10 Jan 202212 Jan 2022
https://www.nldl.org/

Workshop

WorkshopNorthern Lights Deep Learning Workshop
Country/TerritoryNorway
CityTromsø
Period10/01/202212/01/2022
Internet address

Bibliographical note

This work has been funded by the Independent ResearchFund Denmark under case number 9131-00128B.

Fingerprint

Dive into the research topics of 'Re-Identification of Giant Sunfish using Keypoint Matching'. Together they form a unique fingerprint.

Cite this