Next Article in Journal
Design of a Measurement System for Six-Degree-of-Freedom Geometric Errors of a Linear Guide of a Machine Tool
Previous Article in Journal
Strategy for Fault Diagnosis on Train Plug Doors Using Audio Sensors
Article

Coarse-to-Fine Adaptive People Detection for Video Sequences by Maximizing Mutual Information

Video Processing and Understanding Lab (VPULab), Universidad Autónoma de Madrid, 28049 Madrid, Spain
*
Author to whom correspondence should be addressed.
This paper is extension version of the conference paper: García-Martín, A.; SanMiguel, J.C. Adaptive people detection based on cross-correlation maximization. In Proceedins of the 2017 IEEE International Conference on Image Processing (ICIP), Beijing, China, 17–20 September 2017.
These authors contributed equally to this work.
Sensors 2019, 19(1), 4; https://doi.org/10.3390/s19010004
Received: 1 November 2018 / Revised: 29 November 2018 / Accepted: 17 December 2018 / Published: 20 December 2018
(This article belongs to the Section Intelligent Sensors)
Applying people detectors to unseen data is challenging since patterns distributions, such as viewpoints, motion, poses, backgrounds, occlusions and people sizes, may significantly differ from the ones of the training dataset. In this paper, we propose a coarse-to-fine framework to adapt frame by frame people detectors during runtime classification, without requiring any additional manually labeled ground truth apart from the offline training of the detection model. Such adaptation make use of multiple detectors mutual information, i.e., similarities and dissimilarities of detectors estimated and agreed by pair-wise correlating their outputs. Globally, the proposed adaptation discriminates between relevant instants in a video sequence, i.e., identifies the representative frames for an adaptation of the system. Locally, the proposed adaptation identifies the best configuration (i.e., detection threshold) of each detector under analysis, maximizing the mutual information to obtain the detection threshold of each detector. The proposed coarse-to-fine approach does not require training the detectors for each new scenario and uses standard people detector outputs, i.e., bounding boxes. The experimental results demonstrate that the proposed approach outperforms state-of-the-art detectors whose optimal threshold configurations are previously determined and fixed from offline training data. View Full-Text
Keywords: people detection; detector adaptation; pair-wise correlation; thresholds; entropy; coarse-to-fine adaptation people detection; detector adaptation; pair-wise correlation; thresholds; entropy; coarse-to-fine adaptation
Show Figures

Figure 1

MDPI and ACS Style

García-Martín, Á.; SanMiguel, J.C.; Martínez, J.M. Coarse-to-Fine Adaptive People Detection for Video Sequences by Maximizing Mutual Information . Sensors 2019, 19, 4. https://doi.org/10.3390/s19010004

AMA Style

García-Martín Á, SanMiguel JC, Martínez JM. Coarse-to-Fine Adaptive People Detection for Video Sequences by Maximizing Mutual Information . Sensors. 2019; 19(1):4. https://doi.org/10.3390/s19010004

Chicago/Turabian Style

García-Martín, Álvaro, Juan C. SanMiguel, and José M. Martínez 2019. "Coarse-to-Fine Adaptive People Detection for Video Sequences by Maximizing Mutual Information " Sensors 19, no. 1: 4. https://doi.org/10.3390/s19010004

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop