AquaSAM: Underwater Image Foreground Segmentation
This work addresses segmentation challenges for underwater imaging applications, but it is incremental as it fine-tunes an existing model on new data.
The paper tackles the problem of underwater image segmentation by adapting the Segment Anything Model (SAM) to this domain, resulting in AquaSAM, which improves average Dice Similarity Coefficient by 7.13% and mIoU by 8.27% over the default SAM on tasks like coral reefs.
The Segment Anything Model (SAM) has revolutionized natural image segmentation, nevertheless, its performance on underwater images is still restricted. This work presents AquaSAM, the first attempt to extend the success of SAM on underwater images with the purpose of creating a versatile method for the segmentation of various underwater targets. To achieve this, we begin by classifying and extracting various labels automatically in SUIM dataset. Subsequently, we develop a straightforward fine-tuning method to adapt SAM to general foreground underwater image segmentation. Through extensive experiments involving eight segmentation tasks like human divers, we demonstrate that AquaSAM outperforms the default SAM model especially at hard tasks like coral reefs. AquaSAM achieves an average Dice Similarity Coefficient (DSC) of 7.13 (%) improvement and an average of 8.27 (%) on mIoU improvement in underwater segmentation tasks.