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

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

22 Apr 2019arXiv:1904.09730archive 2025-07-28

Youngwan Lee, Joong-won Hwang, Sangrok Lee, Yuseok Bae, Jongyoul Park

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 2x faster speed and the energy consumptions are reduced by 1.6x - 4.1x. 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.

PaperPDFCode

In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

2anchao/VovJpu mentioned on GitHubpytorchSyntology: not harvested report
Dandelight/VoVNet-PaddlePaddle mentioned on GitHubpaddleSyntology: not harvested report
neilctwu/FCOS-pytorch_Simplified mentioned on GitHubpytorchSyntology: not harvested report
osmr/imgclsmob mentioned on GitHubmxnetMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to ResNeSt: Split-Attention Networks (arXiv:2004.08955; that paper's own run record: 28 ran (of which 0 constructed an object rather than computing a result; 21 with no instrument failure: 0 honoured, 0 violated, 21 with no contract checked; 7 where Syntology's instrument failed) · 20 unverified); nothing checks that they implement this paper's method. report
stigma0617/VoVNet-DeepLabV3 mentioned on GitHubpytorchSyntology: not harvested report
stigma0617/VoVNet.pytorch mentioned on GitHubpytorchSyntology: not harvested report
stigma0617/maskrcnn-benchmark-vovnet mentioned on GitHubpytorchMITSyntology: not harvested report
suvasis/birdnet2cs231n mentioned on GitHubpytorchApache-2.0Syntology: not harvested report
vov-net/VoVNet-Detectron mentioned on GitHubpytorchSyntology: not harvested report
vov-net/VoVNet-FCOS mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
youngwanLEE/VoVNet-RefineDet mentioned on GitHubSyntology: not harvested report
youngwanLEE/vovnet-detectron2 mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Object DetectionReal-Time Object DetectionSemantic SegmentationSmall Object Detectionobject-detection

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Instance Segmentation COCO test-dev VoVNetV1-57 mask AP 40.8% #66 of 112 Archive leaderboard report
Instance Segmentation COCO test-dev VoVNetV1-39 mask AP 39.7% #75 of 112 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Introduced by this paper: One-Shot Aggregation

1x1 ConvolutionAverage PoolingBatch NormalizationBottleneck Residual BlockConcatenated Skip ConnectionConvolutionGlobal Average PoolingKaiming InitializationMask R-CNNMax PoolingOne-Shot AggregationRPNReLUResidual BlockResidual ConnectionRoIAlignSPEEDSoftmaxStep DecayVoVNet

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections