Papers › HoughNet: Integrating near and long-range evidence for bottom-up object detection

HoughNet: Integrating near and long-range evidence for bottom-up object detection

5 Jul 2020ECCV 2020 8arXiv:2007.02355archive 2025-07-28

Nermin Samet, Samet Hicsonmez, Emre Akbas

This paper presents HoughNet, a one-stage, anchor-free, voting-based, bottom-up object detection method. Inspired by the Generalized Hough Transform, HoughNet determines the presence of an object at a certain location by the sum of the votes cast on that location. Votes are collected from both near and long-distance locations based on a log-polar vote field. Thanks to this voting mechanism, HoughNet is able to integrate both near and long-range, class-conditional evidence for visual recognition, thereby generalizing and enhancing current object detection methodology, which typically relies on only local evidence. On the COCO dataset, HoughNet's best model achieves 46.4 AP (and 65.1 AP₅₀), performing on par with the state-of-the-art in bottom-up object detection and outperforming most major one-stage and two-stage methods. We further validate the effectiveness of our proposal in another task, namely, "labels to photo" image generation by integrating the voting module of HoughNet to two different GAN models and showing that the accuracy is significantly improved in both cases. Code is available at https://github.com/nerminsamet/houghnet.

PaperPDFConference PDFCode

Code

giddyyupp/coco-minitrain officialmentioned in papermentioned on GitHubpytorch report
nerminsamet/houghnet officialmentioned in papermentioned on GitHubpytorch 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.

Code Syntology ran Syntology

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

Tasks

Image GenerationObjectObject Detectionobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Object Detection COCO minival HoughNet (HG-104, MS) AP50 64.6 #108 of 220 Archive leaderboard report
Object Detection COCO minival HoughNet (HG-104, MS) AP75 50.3 #108 of 220 Archive leaderboard report
Object Detection COCO minival HoughNet (HG-104, MS) APL 59.7 #108 of 220 Archive leaderboard report
Object Detection COCO minival HoughNet (HG-104, MS) APM 48.8 #108 of 220 Archive leaderboard report
Object Detection COCO minival HoughNet (HG-104, MS) APS 30.0 #108 of 220 Archive leaderboard report
Object Detection COCO minival HoughNet (HG-104, MS) box AP 46.1 #108 of 220 Archive leaderboard report
Object Detection COCO minival HoughNet (HG-104) AP50 62.2 #148 of 220 Archive leaderboard report
Object Detection COCO minival HoughNet (HG-104) AP75 46.9 #148 of 220 Archive leaderboard report
Object Detection COCO minival HoughNet (HG-104) APL 55.8 #148 of 220 Archive leaderboard report
Object Detection COCO minival HoughNet (HG-104) APM 47.6 #148 of 220 Archive leaderboard report
Object Detection COCO minival HoughNet (HG-104) APS 25.5 #148 of 220 Archive leaderboard report
Object Detection COCO minival HoughNet (HG-104) box AP 43.0 #148 of 220 Archive leaderboard report
Object Detection COCO test-dev HoughNet (MS) AP50 65.1 #126 of 225 Archive leaderboard report
Object Detection COCO test-dev HoughNet (MS) AP75 50.7 #126 of 225 Archive leaderboard report
Object Detection COCO test-dev HoughNet (MS) APL 58.1 #126 of 225 Archive leaderboard report
Object Detection COCO test-dev HoughNet (MS) APM 48.5 #126 of 225 Archive leaderboard report
Object Detection COCO test-dev HoughNet (MS) APS 29.1 #126 of 225 Archive leaderboard report
Object Detection COCO test-dev HoughNet (MS) box mAP 46.4 #126 of 225 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

1x1 ConvolutionAdamAverage PoolingBatch NormalizationBottleneck Residual BlockConcatenated Skip ConnectionConvolutionCycle Consistency LossDEXTRDeformable ConvolutionDilated ConvolutionDropoutExtremeNetGAN Least Squares LossGlobal Average PoolingHourglass ModuleInstance NormalizationKaiming InitializationMax PoolingPatchGANPix2PixPyramid Pooling ModuleReLUResidual BlockResidual ConnectionSigmoid ActivationSoft-NMSStacked Hourglass NetworkTanh Activation

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