A detection pre-training framework that dynamically rebalances rare classes and adds dual reconstruction improves tail-class AP on COCO and LVIS by about 0.5 to 1.5 points in most configurations.
Siamese DETR
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Long-Tailed Object Detection Pre-training: Dynamic Rebalancing Contrastive Learning with Dual Reconstruction
A detection pre-training framework that dynamically rebalances rare classes and adds dual reconstruction improves tail-class AP on COCO and LVIS by about 0.5 to 1.5 points in most configurations.