000148683 001__ 148683
000148683 005__ 20190509132325.0
000148683 0247_ $$2doi$$a10.5075/epfl-thesis-4746
000148683 02470 $$2urn$$aurn:nbn:ch:bel-epfl-thesis4746-8
000148683 02471 $$2nebis$$a6051007
000148683 037__ $$aTHESIS
000148683 041__ $$aeng
000148683 088__ $$a4746
000148683 245__ $$aLearning Pose Invariant and Covariant Classifiers from Image Sequences
000148683 269__ $$a2010
000148683 260__ $$bEPFL$$c2010$$aLausanne
000148683 300__ $$a132
000148683 336__ $$aTheses
000148683 520__ $$aObject tracking and detection over a wide range of viewpoints is a long-standing problem in Computer Vision. Despite significant advance in wide-baseline sparse interest point matching and development of robust dense feature models, it remains a largely open problem. Moreover, abundance of low cost mobile platforms and novel application areas, such as real-time Augmented Reality, constantly push the performance limits of existing methods. There is a need to modify and adapt these to meet more stringent speed and capacity requirements. In this thesis, we aim to overcome the difficulties due to the multi-view nature of the object detection task. We significantly improve upon existing statistical keypoint matching algorithms to perform fast and robust recognition of image patches independently of object pose. We demonstrate this on various 2D and 3D datasets. The statistical keypoint matching approaches require massive amounts of training data covering a wide range of viewpoints. We have developed a weakly supervised algorithm to greatly simplify their training for 3D objects. We also integrate this algorithm in a 3D tracking-by-detection system to perform real-time Augmented Reality. Finally, we extend the use of a large training set with smooth viewpoint variation to category-level object detection. We introduce a new dataset with continuous pose annotations which we use to train pose estimators for objects of a single category. By using these estimators' output to select pose specific classifiers, our framework can simultaneously localize objects in an image and recover their pose. These decoupled pose estimation and classification steps yield improved detection rates. Overall, we rely on image and video sequences to train classifiers that can either operate independently of the object pose or recover the pose parameters explicitly. We show that in both cases our approaches mitigate the effects of viewpoint changes and improve the recognition performance.
000148683 6531_ $$acomputer vision
000148683 6531_ $$akeypoint recognition
000148683 6531_ $$anaive Bayes
000148683 6531_ $$atracking-by-detection
000148683 6531_ $$aobject detection
000148683 6531_ $$amulti-view
000148683 6531_ $$avision par ordinateur
000148683 6531_ $$areconnaissance de points d'intérêt
000148683 6531_ $$asuivi-par-détection
000148683 6531_ $$adétection d'objets
000148683 6531_ $$amulti-vue
000148683 700__ $$0242714$$g167953$$aÖzuysal, Mustafa
000148683 720_2 $$aFua, Pascal$$edir.$$g112366$$0240252
000148683 720_2 $$aLepetit, Vincent$$edir.$$g149007$$0240235
000148683 8564_ $$uhttps://infoscience.epfl.ch/record/148683/files/EPFL_TH4746.pdf$$zTexte intégral / Full text$$s37886773$$yTexte intégral / Full text
000148683 909C0 $$xU10659$$0252087$$pCVLAB
000148683 909CO $$pthesis-bn2018$$pDOI$$pIC$$ooai:infoscience.tind.io:148683$$qDOI2$$qGLOBAL_SET$$pthesis
000148683 918__ $$dEDIC2005-2015$$cISIM$$aIC
000148683 919__ $$aCVLAB
000148683 920__ $$b2010
000148683 970__ $$a4746/THESES
000148683 973__ $$sPUBLISHED$$aEPFL
000148683 980__ $$aTHESIS