Matches in SemOpenAlex for { <https://semopenalex.org/work/W3138360141> ?p ?o ?g. }
- W3138360141 abstract "Accurate pedestrian classification and localization have received considerable attention due to their wide applications such as security monitoring, autonomous driving, etc. Although pedestrian detectors have made great progress in recent years, the fixed Intersection over Union (IoU) based assignment-regression manner still limits their performance. Two main factors are responsible for this: 1) the IoU threshold faces a dilemma that a lower one will result in more false positives, while a higher one will filter out the matched positives; 2) the IoU-based GT-Proposal assignment suffers from the inconsistent supervision problem that spatially adjacent proposals with similar features are assigned to different ground-truth boxes, which means some very similar proposals may be forced to regress towards different targets, and thus confuses the bounding-box regression when predicting the location results. In this paper, we first put forward the question that textbf{Regression Direction} would affect the performance for pedestrian detection. Consequently, we address the weakness of IoU by introducing one geometric sensitive search algorithm as a new assignment and regression metric. Different from the previous IoU-based textbf{one-to-one} assignment manner of one proposal to one ground-truth box, the proposed method attempts to seek a reasonable matching between the sets of proposals and ground-truth boxes. Specifically, we boost the MR-FPPI under R$_{75}$ by 8.8% on Citypersons dataset. Furthermore, by incorporating this method as a metric into the state-of-the-art pedestrian detectors, we show a consistent improvement." @default.
- W3138360141 created "2021-03-29" @default.
- W3138360141 creator A5022574607 @default.
- W3138360141 creator A5023787090 @default.
- W3138360141 creator A5033797804 @default.
- W3138360141 creator A5034695330 @default.
- W3138360141 creator A5068911982 @default.
- W3138360141 date "2021-03-18" @default.
- W3138360141 modified "2023-09-26" @default.
- W3138360141 title "Which to Match? Selecting Consistent GT-Proposal Assignment for Pedestrian Detection" @default.
- W3138360141 cites W1514535095 @default.
- W3138360141 cites W1861492603 @default.
- W3138360141 cites W1970850659 @default.
- W3138360141 cites W1976818984 @default.
- W3138360141 cites W2031454541 @default.
- W3138360141 cites W2041059661 @default.
- W3138360141 cites W2081021369 @default.
- W3138360141 cites W2085805060 @default.
- W3138360141 cites W2088049833 @default.
- W3138360141 cites W2096349671 @default.
- W3138360141 cites W2097324787 @default.
- W3138360141 cites W2098699644 @default.
- W3138360141 cites W2107775979 @default.
- W3138360141 cites W2161969291 @default.
- W3138360141 cites W2168356304 @default.
- W3138360141 cites W2194775991 @default.
- W3138360141 cites W2200528286 @default.
- W3138360141 cites W2407521645 @default.
- W3138360141 cites W2490270993 @default.
- W3138360141 cites W2497039038 @default.
- W3138360141 cites W2530479064 @default.
- W3138360141 cites W2531915888 @default.
- W3138360141 cites W2549139847 @default.
- W3138360141 cites W2565639579 @default.
- W3138360141 cites W2570343428 @default.
- W3138360141 cites W2578555672 @default.
- W3138360141 cites W2579152745 @default.
- W3138360141 cites W2592268053 @default.
- W3138360141 cites W2594507094 @default.
- W3138360141 cites W2604243686 @default.
- W3138360141 cites W2736410039 @default.
- W3138360141 cites W2743473392 @default.
- W3138360141 cites W2792824754 @default.
- W3138360141 cites W2809784273 @default.
- W3138360141 cites W2810862788 @default.
- W3138360141 cites W2883363148 @default.
- W3138360141 cites W2894820835 @default.
- W3138360141 cites W2896540732 @default.
- W3138360141 cites W2903748531 @default.
- W3138360141 cites W2935837427 @default.
- W3138360141 cites W2943081120 @default.
- W3138360141 cites W2950167387 @default.
- W3138360141 cites W2951548327 @default.
- W3138360141 cites W2962766617 @default.
- W3138360141 cites W2962850098 @default.
- W3138360141 cites W2963091558 @default.
- W3138360141 cites W2963179609 @default.
- W3138360141 cites W2963315052 @default.
- W3138360141 cites W2963318220 @default.
- W3138360141 cites W2963399829 @default.
- W3138360141 cites W2963403868 @default.
- W3138360141 cites W2963404857 @default.
- W3138360141 cites W2963420686 @default.
- W3138360141 cites W2963681621 @default.
- W3138360141 cites W2963722952 @default.
- W3138360141 cites W2963786238 @default.
- W3138360141 cites W2963998989 @default.
- W3138360141 cites W2964044605 @default.
- W3138360141 cites W2964211037 @default.
- W3138360141 cites W2964241181 @default.
- W3138360141 cites W2970575838 @default.
- W3138360141 cites W2974603289 @default.
- W3138360141 cites W2976098368 @default.
- W3138360141 cites W2976700867 @default.
- W3138360141 cites W2982770724 @default.
- W3138360141 cites W2990075400 @default.
- W3138360141 cites W2990946490 @default.
- W3138360141 cites W2991833700 @default.
- W3138360141 cites W2995649684 @default.
- W3138360141 cites W2997344726 @default.
- W3138360141 cites W2997747012 @default.
- W3138360141 cites W3001652308 @default.
- W3138360141 cites W3034638324 @default.
- W3138360141 cites W3035323039 @default.
- W3138360141 cites W3042641282 @default.
- W3138360141 cites W3095638603 @default.
- W3138360141 cites W3106250896 @default.
- W3138360141 cites W603908379 @default.
- W3138360141 cites W639708223 @default.
- W3138360141 doi "https://doi.org/10.48550/arxiv.2103.10091" @default.
- W3138360141 hasPublicationYear "2021" @default.
- W3138360141 type Work @default.
- W3138360141 sameAs 3138360141 @default.
- W3138360141 citedByCount "0" @default.
- W3138360141 crossrefType "posted-content" @default.
- W3138360141 hasAuthorship W3138360141A5022574607 @default.
- W3138360141 hasAuthorship W3138360141A5023787090 @default.
- W3138360141 hasAuthorship W3138360141A5033797804 @default.
- W3138360141 hasAuthorship W3138360141A5034695330 @default.
- W3138360141 hasAuthorship W3138360141A5068911982 @default.