Automatic Coral Reef Annotation, Localization and Pixel-wise Parsing Using Mask R-CNN

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F21%3A43962958" target="_blank" >RIV/49777513:23520/21:43962958 - isvavai.cz</a>
Result on the web
<a href="http://ceur-ws.org/Vol-2936/paper-115.pdf" target="_blank" >http://ceur-ws.org/Vol-2936/paper-115.pdf</a>
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Automatic Coral Reef Annotation, Localization and Pixel-wise Parsing Using Mask R-CNN
Original language description
This paper describes the methods that were used for annotation, localization and pixel-wise parsing of the coral reefs from underwater images. The proposed system achieved competitive results in the third edition of ImageCLEFcoral 2021 challenge. Specifically, in case of annotation and localization task achieved mean average precision with Intersection over Union (IoU) greater that 0.5 (mAP@0.5) 0.121 and in case of pixel-wise parsing task achieved mAP@0.5 0.075 on the test set. The proposed method is based on Mask R-CNN object detection and segmentation framework with online data augmentations.
Czech name
—
Czech description
—

Publication year
2021
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
Working Notes of CLEF 2021 - Conference and Labs of the Evaluation Forum
ISBN
—
ISSN
1613-0073
e-ISSN
—
Number of pages
6
Pages from-to
1359-1364
Publisher name
CEUR-WS
Place of publication
Bucharest
Event location
Bucharest, Romania (virtual)
Event date
Sep 21, 2021
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—

Similar results(10)