A
Amir Shahroudy
Researcher at Nanyang Technological University
Publications - 5
Citations - 1550
Amir Shahroudy is an academic researcher from Nanyang Technological University. The author has contributed to research in topics: Feature extraction & Deep learning. The author has an hindex of 4, co-authored 5 publications receiving 1510 citations.
Papers
More filters
Posted Content
Recent Advances in Convolutional Neural Networks
Jiuxiang Gu,Zhenhua Wang,Jason Kuen,Lianyang Ma,Amir Shahroudy,Bing Shuai,Ting Liu,Xingxing Wang,Li Wang,Gang Wang,Jianfei Cai,Tsuhan Chen +11 more
TL;DR: This paper details the improvements of CNN on different aspects, including layer design, activation function, loss function, regularization, optimization and fast computation, and introduces various applications of convolutional neural networks in computer vision, speech and natural language processing.
Journal ArticleDOI
Multimodal Multipart Learning for Action Recognition in Depth Videos
TL;DR: This work proposes a joint sparse regression based learning method which utilizes the structured sparsity to model each action as a combination of multimodal features from a sparse set of body parts, to represent dynamics and appearance of parts.
Proceedings ArticleDOI
SSNet: Scale Selection Network for Online 3D Action Prediction
TL;DR: This paper focuses on online action prediction in streaming 3D skeleton sequences and proposes a novel window scale selection scheme to make the network focus on the performed part of the ongoing action and try to suppress the noise from the previous actions at each time step.
Proceedings ArticleDOI
Multi-modal feature fusion for action recognition in RGB-D sequences
TL;DR: This paper proposed a new hierarchical bag-of-words feature fusion technique based on multi-view structured spar-sity learning to fuse atomic features from RGB and skeletons for the task of action recognition.
DissertationDOI
Activity recognition in depth videos
TL;DR: This thesis proposes a joint sparse regression based learning method which utilizes the structured sparsity to model each action as a combination of multimodal features from a sparse set of body parts, and proposes a new deep autoencoder-based correlation-independence factorization network to separate input multi-modality signals into a hierarchy of extracted components.