Patch ensembles anchored to predicted lateral lines improve salmon re-identification mAP from 0.609 to 0.860 in cross-camera tests over full-image baselines.
Py- torch metric learning.arXiv preprint arXiv:2008.09164
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
Self-supervised ReLU networks form substantially fewer linear regions than supervised models for comparable accuracy, with contrastive methods rapidly expanding regions and self-distillation consolidating them, enabling early geometric detection of representation collapse.
citing papers explorer
-
Patch Ensembles for Robust Salmon Re-Identification with Weak Trajectory Labels
Patch ensembles anchored to predicted lateral lines improve salmon re-identification mAP from 0.609 to 0.860 in cross-camera tests over full-image baselines.
-
Complexity of Linear Regions in Self-supervised Deep ReLU Networks
Self-supervised ReLU networks form substantially fewer linear regions than supervised models for comparable accuracy, with contrastive methods rapidly expanding regions and self-distillation consolidating them, enabling early geometric detection of representation collapse.