1 / 15

Evaluation of Manually Created Ground Truth for Multi-view People Localization

Evaluation of Manually Created Ground Truth for Multi-view People Localization. Ákos Kiss, Tamás Szirányi Distributed Events Analysis Research Laboratory kiss.akos @ sztaki.mta.hu Sziranyi.tamas@sztaki.mta.hu. Multi-View People Localization Overview. Problem definition Motivation

drea
Download Presentation

Evaluation of Manually Created Ground Truth for Multi-view People Localization

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Evaluation of ManuallyCreatedGroundTruthforMulti-viewPeopleLocalization Ákos Kiss, Tamás Szirányi DistributedEventsAnalysis Research Laboratory kiss.akos@sztaki.mta.hu Sziranyi.tamas@sztaki.mta.hu

  2. Multi-ViewPeopleLocalization Overview • Problemdefinition • Motivation • Ourframework • Contribution • Referencegroundtruth • Evaluating human observers • Summary

  3. Multi-ViewPeopleLocalization Problemdefinition • Multiplecameras monitoring an area • Overlappingfield of view • Additionalspatialinformation • Pixelscorrespondtolinesinspace • Linesintersectwhereobjectlies • Triangulation • Groundtruthrequired

  4. Multi-ViewPeopleLocalization Problemdefinition • Creatinggroundtruth • Manually • Positioninimages • Boundingboxes • Positioninreferencespace • Parametricsurface • True 3D positions • Automatically • ToFsensors • Kinect • Lidar - Expensivesolutions - Occlusion is still a problem

  5. Multi-ViewPeopleLocalization Motivation • Groundtruthusuallytakenforgranted • Human makemistakes • Poorgroundtruthleadstoinvalidalgorithmevaluation • Generatinggroundtruth is timeconsuming • Experts’ time is expensive • Evaluating human observers • 9 subjects (6 laymen, 3 withdomainknowledge)

  6. Multi-ViewPeopleLocalization Ourframework • GUI – allviewsvisible • Locationbytriangulation • Mark inany (atleast 2) views • Triangulation • Feedback (savelocationonlyifcorrect) • Localizingpersonbyperson • Head • Feet • Skipfootifnotvisiblein 2 views

  7. Multi-ViewPeopleLocalization Ourframework • Localizingbylinearoptimization • p is onbothlines ( axis): • Linespracticallyneverintersect • Pseudoinverse (minimalizes SSE) • Validatingwithknownsurface • Planarground is reconstructedprecisely

  8. Multi-ViewPeopleLocalization Referencegroundtruth • Several „groundtruth” createdbysubjects • Combininginformationfrommultiplegroundtruth • Peoplearelocalizedbyonly a subset of subjects • Positionsarenoisy • Automatingprocess • Match peoplebyhead (body) location • Matchingfeet (ordermightdiffer) • Filteringnoise (weighted center of locations) • More reliableresult: referencegroundtruth

  9. Multi-ViewPeopleLocalization Evaluating human observers • Errors (precision) • Accuracy (locationerror) • Recall (missedpeople) • Temporalanalysis

  10. Multi-ViewPeopleLocalization Evaluating human observers • Errors (precision) • Lowerrorrate • Typicalerrors • Parallel lines (mark near camera) • Mix uppeople • Accuracy (locationerror) • Recall (missedpeople) • Temporalanalysis

  11. Multi-ViewPeopleLocalization Evaluating human observers • Errors (precision) • Accuracy (locationerror) • Synchronizationerror • Lowdeviation • Outliersuppression • Iterative • Changeweight of points • Recall (missedpeople) • Temporalanalysis

  12. Multi-ViewPeopleLocalization Evaluating human observers • Errors (precision) • Accuracy (locationerror) • Recall (missedpeople) • Verylowrecall • Expertsarenotbetter • Temporalanalysis feet = headrecall line recallvalues of laymen (blue) and experts (green)

  13. Multi-ViewPeopleLocalization Evaluating human observers • Errors (precision) • Accuracy (locationerror) • Recall (missedpeople) • Temporalanalysis • Shortexperiment • Fewsubjects • Less outliersforexperts

  14. Multi-ViewPeopleLocalization Summary • Typical human observer • Highprecision • Highaccuracy • Lowrecall • Generatingthistype of groundtruthrequiresmuchattention • Generating more reliablereferencegroundtruthcan be automated

  15. ThankYou!

More Related