A new bottleneck module that downsamples then upsamples feature maps inside a lightweight CNN block improves ImageNet accuracy by up to 6.6% over MobileNetV2 at comparable FLOPs.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
HBONet: Harmonious Bottleneck on Two Orthogonal Dimensions
A new bottleneck module that downsamples then upsamples feature maps inside a lightweight CNN block improves ImageNet accuracy by up to 6.6% over MobileNetV2 at comparable FLOPs.