Mostra el registre d'ítem simple
Disentangling motion, foreground and background features in videos
dc.contributor | Torres Viñals, Jordi |
dc.contributor.author | Lin, Xunyu |
dc.date.accessioned | 2018-01-26T09:00:21Z |
dc.date.available | 2018-01-26T09:00:21Z |
dc.date.issued | 2017 |
dc.identifier.uri | http://hdl.handle.net/2117/113234 |
dc.description.abstract | This paper instroduces an unsupervised framework to extract semantically rich features for video representation. Inspired by how the human visual system groups objects based on motion cues, we propose a deep convolutional neural network that disentangles motion, foreground and background information. The proposed architecture consists of a 3D convolutional feature encoder for blocks of 16 frames, which is trained for reconstruction tasks over the first and last frames of the sequence. The model is trained with a fraction of videos from the UCF-101 dataset taking as ground truth the bounding boxes around the activity regions. Qualitative results indicate that the network can successfully update the foreground appearance based on pure-motion features. The benefits of these learned features are shown in a discriminative classification task when compared with a random initialization of the network weights, providing a gain of accuracy above the 10%. |
dc.language.iso | eng |
dc.publisher | Universitat Politècnica de Catalunya |
dc.subject | Àrees temàtiques de la UPC::Informàtica |
dc.subject.lcsh | Artificial intelligence |
dc.subject.lcsh | Video recording |
dc.subject.other | Unsupervised learning |
dc.subject.other | artificial intelligence |
dc.subject.other | video features |
dc.subject.other | action recognition |
dc.title | Disentangling motion, foreground and background features in videos |
dc.title.alternative | Unsupervised video representations learning for activity recognition |
dc.type | Bachelor thesis |
dc.subject.lemac | Intel·ligència artificial |
dc.subject.lemac | Vídeo |
dc.identifier.slug | 128563 |
dc.rights.access | Open Access |
dc.date.updated | 2017-06-30T14:11:12Z |
dc.audience.educationlevel | Grau |
dc.audience.mediator | Facultat d'Informàtica de Barcelona |
dc.audience.degree | GRAU EN ENGINYERIA INFORMÀTICA (Pla 2010) |