Virtual and Real World Adaptation for Pedestrian Detection

Pedestrian detection is of paramount interest for many applications. Most promising detectors rely on discriminatively learnt classifiers, i.e., trained with annotated samples. However, the annotation step is a human intensive and subjective task worth to be minimized. By using virtual worlds we can...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on pattern analysis and machine intelligence 2014-04, Vol.36 (4), p.797-809
Main Authors: Vazquez, David, Lopez, Antonio M., Marin, Javier, Ponsa, Daniel, Geronimo, David
Format: Article
Language:English
Subjects:
Accuracy
Adaptation
Algorithms
Annotations
Applied sciences
Artificial intelligence
Cameras
Classifiers
Computer science
control theory
systems
Computer systems and distributed systems. User interface
data set shift
Detectors
domain adaptation
Exact sciences and technology
General aspects
Image detection
Image Enhancement - methods
Image Interpretation, Computer-Assisted - methods
Image resolution
Interpolation
Machine Learning
Occupational training. Personnel. Work management
Pattern Recognition, Automated - methods
Pattern recognition. Digital image processing. Computational geometry
Pedestrian detection
Pedestrians
Pedestrians - classification
photo-realistic computer animation
Photography - methods
Reproducibility of Results
Sensitivity and Specificity
Software
Subtraction Technique
Testing
Training
User-Computer Interface
Whole Body Imaging - methods
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Pedestrian detection is of paramount interest for many applications. Most promising detectors rely on discriminatively learnt classifiers, i.e., trained with annotated samples. However, the annotation step is a human intensive and subjective task worth to be minimized. By using virtual worlds we can automatically obtain precise and rich annotations. Thus, we face the question: can a pedestrian appearance model learnt in realistic virtual worlds work successfully for pedestrian detection in real-world images? Conducted experiments show that virtual-world based training can provide excellent testing accuracy in real world, but it can also suffer the data set shift problem as real-world based training does. Accordingly, we have designed a domain adaptation framework, V-AYLA, in which we have tested different techniques to collect a few pedestrian samples from the target domain (real world) and combine them with the many examples of the source domain (virtual world) in order to train a domain adapted pedestrian classifier that will operate in the target domain. V-AYLA reports the same detection accuracy than when training with many human-provided pedestrian annotations and testing with real-world images of the same domain. To the best of our knowledge, this is the first work demonstrating adaptation of virtual and real worlds for developing an object detector.
ISSN:0162-8828
1939-3539
2160-9292
DOI:10.1109/TPAMI.2013.163