The developed tool allows a human operator to annotate on a file all pedestrians in a previously acquired video sequence. A similar file is produced by the algorithm being tested using the same annotation engine. A matching rule has been established to validate the association between items of the two files. For each frame a statistical analyzer extracts the number of mis-detections, both positive and negative, and correct detections. Using these data, statistics about the algorithm behavior are computed with the aim of tuning parameters and pointing out recongnition weaknesses in particular situations.

Tool for Vision based Pedestrian Detection Performance Evaluation / Bertozzi, Massimo; Broggi, Alberto; Grisleri, Paolo; Amos, Tibaldi; MICHAEL DEL, Rose. - (2004). (Intervento presentato al convegno Intelligent Vehicles Symposium 2004) [10.1109/IVS.2004.1336484].

Tool for Vision based Pedestrian Detection Performance Evaluation

BERTOZZI, Massimo;BROGGI, Alberto;GRISLERI, Paolo;
2004-01-01

Abstract

The developed tool allows a human operator to annotate on a file all pedestrians in a previously acquired video sequence. A similar file is produced by the algorithm being tested using the same annotation engine. A matching rule has been established to validate the association between items of the two files. For each frame a statistical analyzer extracts the number of mis-detections, both positive and negative, and correct detections. Using these data, statistics about the algorithm behavior are computed with the aim of tuning parameters and pointing out recongnition weaknesses in particular situations.
2004
0780383109
Tool for Vision based Pedestrian Detection Performance Evaluation / Bertozzi, Massimo; Broggi, Alberto; Grisleri, Paolo; Amos, Tibaldi; MICHAEL DEL, Rose. - (2004). (Intervento presentato al convegno Intelligent Vehicles Symposium 2004) [10.1109/IVS.2004.1336484].
File in questo prodotto:
File Dimensione Formato  
iv2004-annotation.pdf

non disponibili

Tipologia: Documento in Post-print
Licenza: Creative commons
Dimensione 480.58 kB
Formato Adobe PDF
480.58 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11381/1440386
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 23
  • ???jsp.display-item.citation.isi??? 10
social impact