PlankFormer uses MAE-pretrained ViT backbones and pseudo community image synthesis to achieve higher-precision plankton instance segmentation than Mask R-CNN in debris-heavy scenes while needing fewer manual annotations.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
PlankFormer: Robust Plankton Instance Segmentation via MAE-Pretrained Vision Transformers and Pseudo Community Image Generation
PlankFormer uses MAE-pretrained ViT backbones and pseudo community image synthesis to achieve higher-precision plankton instance segmentation than Mask R-CNN in debris-heavy scenes while needing fewer manual annotations.