An Energy and GPU-Computation Efficient Backbone Network for Real-Time Object Detection

Youngwan LeeJoong-won HwangSangrok LeeYuseok BaeJongyoul Park

   Papers with code   Abstract  PDF

As DenseNet conserves intermediate features with diverse receptive fields by aggregating them with dense connection, it shows good performance on the object detection task. Although feature reuse enables DenseNet to produce strong features with a small number of model parameters and FLOPs, the detector with DenseNet backbone shows rather slow speed and low energy efficiency... (read more)

Benchmarked Models

RANK
MODEL
REPO
CODE RESULT
PAPER RESULT
ε-REPRODUCED
BUILD
1
VoVNet-57
77.7%
--
2
VoVNet-39
76.8%
--