Napredna pretraga

Pregled bibliografske jedinice broj: 474496

Experimental comparison of AdaBoost algorithms applied on Leg Detection with Different Range Sensor Setups


Jurić-Kavelj, Srećko; Petrović, Ivan
Experimental comparison of AdaBoost algorithms applied on Leg Detection with Different Range Sensor Setups // 19th International Workshop on Robotics in Alpe-Adria-Danube Region – RAAD 2010 / Anikó Szakál (ur.).
Budimpešta: IEEE, 2010. str. 267-272 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)


Naslov
Experimental comparison of AdaBoost algorithms applied on Leg Detection with Different Range Sensor Setups

Autori
Jurić-Kavelj, Srećko ; Petrović, Ivan

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
19th International Workshop on Robotics in Alpe-Adria-Danube Region – RAAD 2010 / Anikó Szakál - Budimpešta : IEEE, 2010, 267-272

ISBN
978-1-4244-6884-3

Skup
19th IEEE International Workshop on Robotics in Alpe-Adria-Danube Region

Mjesto i datum
Budimpešta, Mađarska, 23-25.06.2010.

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
AdaBoost; 2D range; legs; SICK; Hokuyo

Sažetak
When tracking people or other moving objects with a mobile robot, detection is the first and most critical step. At first most researchers focused on the tracking algorithms, but recently AdaBoost (supervised machine learning technique) was used for people legs detection in 2D range data. The results are promising, but it is unclear if the obtained classifier could be used on the data from another sensor. As it would be a huge inconvenience having to train a classifier for every sensor (setup), we set out to find if, and when is a classifier trained on one sensor setup transferable to another sensor setup. We tested two sensors in five different setups. In total, we acquired 2455 range scans. Experiments showed that the classifier trained on noisier sensor data performed better at classification of data coming from other sensor setups. Classifiers trained on less noisy data were shown to be overconfident, and performed poorly on noisy data. Furthermore, experiments showed that classifiers learned on ten times smaller datasets performed as good as classifiers trained on larger datasets. Since AdaBoost is a supervised learning technique, obtaining same classifier efficiency with significantly smaller dataset means less hand labeling of the data for the same results.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Računarstvo, Temeljne tehničke znanosti



POVEZANOST RADA


Projekt / tema
036-0361621-3012 - Napredne strategije upravljanja i estimacije u složenim sustavima (Nedjeljko Perić, )
036-0363078-3018 - Upravljanje mobilnim robotima i vozilima u nepoznatim i dinamičkim okruženjima (Ivan Petrović, )

Ustanove
Fakultet elektrotehnike i računarstva, Zagreb