DSS-USOD decomposes underwater image features into boundary-sensitive and region-coherent branches with a spatial coordination module and cooperative supervision for improved salient object detection under degradations.
A convnet for the 2020s,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
DualOpt decouples optimization by using real-time layer-wise weight decay for scratch training and weight rollback for fine-tuning to improve convergence, generalization, and reduce knowledge forgetting.
citing papers explorer
-
Learning Dynamic Structural Specialization for Underwater Salient Object Detection
DSS-USOD decomposes underwater image features into boundary-sensitive and region-coherent branches with a spatial coordination module and cooperative supervision for improved salient object detection under degradations.
-
Neural Network Optimization Reimagined: Decoupled Techniques for Scratch and Fine-Tuning
DualOpt decouples optimization by using real-time layer-wise weight decay for scratch training and weight rollback for fine-tuning to improve convergence, generalization, and reduce knowledge forgetting.