… we explore the subtle cues necessary for such robust viewpoint-independent recognition. … pose and viewpoint, clothing is not always a reliable cue for personrecognition in our dataset. …
… affective state recognition. As affect recognition systems usingmultiplecues and modalities … 2007, based on multiple visual cues (ie, affective head, face and/or body movement), haptic …
M Yang, F Lv, W Xu, Y Gong - 2009 IEEE 12th International …, 2009 - ieeexplore.ieee.org
… multicue in tracking multiplehuman driven by human detections. Given a humandetection … combine specifically devised models based on different cues in this tracker so as to enhance …
Y Liu, N Song, Y Han - Journal of Visual Communication and Image …, 2019 - Elsevier
… -identification. For traditional features, we usemultiple color spaces for feature extraction, and … all of the color features are fused together to improve the precision of the final recognition. …
Q Xie, Z Lu, W Zhou, H Li - IEEE Transactions on Multimedia, 2022 - ieeexplore.ieee.org
… people under multiplecues. We believe that the use of face cues is able to aid in person re-identification in … 2, A4 is a poor result retrieved through body cues; however, the face in the …
D Swangpol, T Chalidabhongse - 제어로봇시스템학회: 학술대회 …, 2005 - 2005.iccas.org
… , track and recognize persons in real-time. Our goal is to develop a personidentification system that can detect, track, and recognize people from video usingmultiplecues such as face, …
B Stenger - Asian conference on computer vision, 2006 - Springer
… Hand gesture recognitionusingmultiscale colour features, hierarchical models and particle … Fast 2D hand tracking with flocks of features and multi-cue integration. In: Workshop on Real…
S Aboutalib, M Veloso - Proceedings of the 6th international joint …, 2007 - dl.acm.org
… non-visual cues to those objects’ identities. We research on a flexible object recognition approach which can use any multiplecues, whether they are visual cues intrinsic to the object or …
… multiplecues from different levels, our approach can better represent group actions and improve the recognition … (2) We propose new discriminative motion features based on the motion …