A training-free pipeline combining SAM3 for masks and transformed DINOv3 embeddings for prototype matching delivers the first reported baseline for fine-grained fungi segmentation across one-shot to few-hundred-shot regimes.
An investigation into whitening loss for self-supervised learning.Advances in Neural Infor- mation Processing Systems, 35:29748–29760
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
-
Training-Free Fine-Grained Semantic Segmentations in Low Data Regimes: A FungiTastic Baseline
A training-free pipeline combining SAM3 for masks and transformed DINOv3 embeddings for prototype matching delivers the first reported baseline for fine-grained fungi segmentation across one-shot to few-hundred-shot regimes.