MAFE R-CNN uses dynamic multi-clue sample selection and a category-aware feature memory to reach 32.7 AP on SODA-D and 35.8 AP on SODA-A, a modest improvement over prior detectors.
Accurate and robust object detection via selective adversarial learning with constraints,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
MAFE R-CNN: Selecting More Samples to Learn Category-aware Features for Small Object Detection
MAFE R-CNN uses dynamic multi-clue sample selection and a category-aware feature memory to reach 32.7 AP on SODA-D and 35.8 AP on SODA-A, a modest improvement over prior detectors.