Détail de l'auteur
Auteur Krystian Mikolajczyk |
Documents disponibles écrits par cet auteur (1)
Ajouter le résultat dans votre panier
Visionner les documents numériques
Affiner la recherche Interroger des sources externes
Tracking-Learning-Detection / Zdenek Kalal in IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI, vol 34 n° 7 (July 2012)
[article]
Titre : Tracking-Learning-Detection Type de document : Article/Communication Auteurs : Zdenek Kalal, Auteur ; Krystian Mikolajczyk, Auteur ; Jiri Matas, Auteur Année de publication : 2012 Article en page(s) : pp 1409 - 1422 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Photogrammétrie numérique
[Termes IGN] apprentissage automatique
[Termes IGN] détection de cible
[Termes IGN] poursuite de cible
[Termes IGN] séquence d'imagesRésumé : (auteur) This paper investigates long-term tracking of unknown objects in a video stream. The object is defined by its location and extent in a single frame. In every frame that follows, the task is to determine the object's location and extent or indicate that the object is not present. We propose a novel tracking framework (TLD) that explicitly decomposes the long-term tracking task into tracking, learning, and detection. The tracker follows the object from frame to frame. The detector localizes all appearances that have been observed so far and corrects the tracker if necessary. The learning estimates the detector's errors and updates it to avoid these errors in the future. We study how to identify the detector's errors and learn from them. We develop a novel learning method (P-N learning) which estimates the errors by a pair of “experts”: (1) P-expert estimates missed detections, and (2) N-expert estimates false alarms. The learning process is modeled as a discrete dynamical system and the conditions under which the learning guarantees improvement are found. We describe our real-time implementation of the TLD framework and the P-N learning. We carry out an extensive quantitative evaluation which shows a significant improvement over state-of-the-art approaches. Numéro de notice : A2012-720 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TPAMI.2011.239 En ligne : https://doi.org/10.1109/TPAMI.2011.239 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=84845
in IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI > vol 34 n° 7 (July 2012) . - pp 1409 - 1422[article]