Aug 2026· Machine Learning for Computational Science and Engineering· Vol 2· 0 citations· 24 references
TL;DR
ProCAMMix accumulates stable per-class attention representations through an exponential moving average (EMA) buffer and introduces a progressive annealing schedule that interpolates from standard CutMix behaviour in early training epochs to fully CAM-guided mixing in later epochs—when CAMs are discriminative.
This work introduces DR.WILSS, an innovative approach to address catastrophic forgetting in continual learning using diffusion-based generative replay, which leverages language clues to guide the diffusion process, employing self-inpainting and regularization techniques to efficiently produce replay data, aiding the le...
Leon Arthur Marx, Francesco Barbato, Matteo Caligiuri et al.· 0 citations
Data augmentation is a standard component of modern semantic segmentation pipelines, but most augmentation techniques allocate transformations uniformly across training samples or adapt to a single difficulty signal such as loss. This ignores the fact that segmentation difficulty is multi-factorial, since ambiguous pre...
O. Arigbabu, Abimbola Ismail Arigbabu· 0 citations
Adversarial erasing (AE) is commonly used to expand class activation maps (CAM) in Weakly Supervised Semantic Segmentation (WSSS), but existing AE methods struggle with semantic drift, background over-activation, and boundary blur due to limited structural awareness. We propose a CAM optimization method that combines c...
Ya-Wen Shi, Xu Chen, Jin-Ping Tang et al.· International Conference on...· 0 citations
A deep mixed data augmentation framework that jointly enhances both the support set and the query set in few-shot image classification and is competitive in few-shot classification.
This work proposes a two-stage framework that introduces SAM2 priors into PS-CD and progressively adapts them to the target task, and establishes a closed-loop optimization process that alternates between pseudo-label refinement and model re-optimization.
Hailong Ning, Hao Wang, Yiming Wang et al.· 0 citations
Aether is introduced, a simple plug-in method that applies diffusion-style random perturbations in the embedding space via controlled alpha-mixing, specifically designed to provide isotropic regularization that remains semantically consistent.
Hyesong Choi, Daeun Kim, Song Park et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
