FSS-1000
PulseAugur coverage of FSS-1000 — every cluster mentioning FSS-1000 across labs, papers, and developer communities, ranked by signal.
-
SADe improves few-shot segmentation with weak annotation cleaning
Researchers have developed SADe, a novel layer designed to improve few-shot segmentation by cleaning up weak support annotations. This method uses sparse autoencoder atom evidence to estimate the reliability of support …
-
REBASE framework enhances in-context segmentation by eliminating background noise · 2 sources tracked
Researchers have introduced REBASE, a novel training-free framework designed to improve in-context segmentation. This method addresses limitations in existing approaches by explicitly suppressing spurious contextual cor…
-
REBASE framework enhances training-free in-context segmentation
Researchers have developed REBASE, a novel framework designed to improve training-free in-context segmentation. This method explicitly suppresses spurious contextual correspondences by identifying and eliminating the lo…