A tool for vision based pedestrian detection performance evaluation
- 8 November 2004
- proceedings article
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 784-789
- https://doi.org/10.1109/ivs.2004.1336484
Abstract
This paper describes a system for evaluating pedestrian detection algorithm results. 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 weak- nesses in particular situations.Keywords
This publication has 4 references indexed in Scilit:
- Shape-based pedestrian detection and localizationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2004
- Stereo-based preprocessing for human shape localization in unstructured environmentsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2004
- Pedestrian detection in infrared imagesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2004
- Tools and techniques for video performance evaluationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002