ETRI-Knowledge Sharing Plaform

논문 검색
Type SCI
Year ~ Keyword


Conference Paper An Energy and GPU-Computation Efficient Backbone Network for Real-Time Object Detection
Cited 257 time in scopus Download 24 time Share share facebook twitter linkedin kakaostory
Youngwan Lee, Joong-Won Hwang, 이상록, Bae Yu Seok, Park Jongyoul
Issue Date
Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) 2019, pp.752-760
Project Code
19HS3400, Development of High Performance Visual Discovery Platform for Realtime and Large-Scale Data Analysis and Prediction, Park Jongyoul
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. We find the linearly increasing input channel by dense connection leads to heavy memory access cost, which causes computation overhead and more energy consumption. To solve the inefficiency of DenseNet, we propose an energy and computation efficient architecture called VoVNet comprised of One-Shot Aggregation (OSA). The OSA not only adopts the strength of DenseNet that represents diversified features with multi receptive fields but also overcomes the inefficiency of dense connection by aggregating all features only once in the last feature maps. To validate the effectiveness of VoVNet as a backbone network, we design both lightweight and large-scale VoVNet and apply them to one-stage and two-stage object detectors. Our VoVNet based detectors outperform DenseNet based ones with 2× faster speed and the energy consumptions are reduced by 1.6× - 4.1×. In addition to DenseNet, VoVNet also outperforms widely used ResNet backbone with faster speed and better energy efficiency. In particular, the small object detection performance has been significantly improved over DenseNet and ResNet.
KSP Keywords
Backbone Network, Detection task, Energy Efficiency, Feature Map, Low energy, Memory access cost, Model parameter, One-stage, Real-time object detection, Receptive field, Two-Stage