Open AccessPosted Content
Convolutional Two-Stream Network Fusion for Video Action Recognition
TLDR
In this paper, a spatial and temporal network can be fused at the last convolution layer without loss of performance, but with a substantial saving in parameters, and furthermore, pooling of abstract convolutional features over spatiotemporal neighbourhoods further boosts performance.Abstract:
Recent applications of Convolutional Neural Networks (ConvNets) for human action recognition in videos have proposed different solutions for incorporating the appearance and motion information. We study a number of ways of fusing ConvNet towers both spatially and temporally in order to best take advantage of this spatio-temporal information. We make the following findings: (i) that rather than fusing at the softmax layer, a spatial and temporal network can be fused at a convolution layer without loss of performance, but with a substantial saving in parameters; (ii) that it is better to fuse such networks spatially at the last convolutional layer than earlier, and that additionally fusing at the class prediction layer can boost accuracy; finally (iii) that pooling of abstract convolutional features over spatiotemporal neighbourhoods further boosts performance. Based on these studies we propose a new ConvNet architecture for spatiotemporal fusion of video snippets, and evaluate its performance on standard benchmarks where this architecture achieves state-of-the-art results.read more
Citations
More filters
Posted Content
POSEidon: Face-from-Depth for Driver Pose Estimation
TL;DR: A new deep learning framework for head localization and pose estimation on depth images and a new Face-from-Depth model for learning image faces from depth are presented, which overcomes all recent state-of-art works in face reconstruction.
Proceedings ArticleDOI
Computer vision approaches based on deep learning and neural networks: Deep neural networks for video analysis of human pose estimation
Eralda Nishani,Betim Cico +1 more
TL;DR: A systematic mapping study investigates existing research about implementations of computer vision approaches based on deep learning algorithms and Convolutional Neural Networks and proposes three different research direction related to: improving existing CNN implementations, using Recurrent Neural Networks (RNNs) for human pose estimation and finally relying on unsupervised learning paradigm to train NNs.
Journal ArticleDOI
Supervised Classification of Multisensor Remotely Sensed Images Using a Deep Learning Framework
TL;DR: A convolutional neural network (CNN)-based method to efficiently combine information from multisensor remotely sensed images for pixel-wise semantic classification and a composite fusion architecture that fuses features throughout the network are presented.
Posted Content
Late Temporal Modeling in 3D CNN Architectures with BERT for Action Recognition
TL;DR: This work replaces the conventional Temporal Global Average Pooling layer at the end of 3D convolutional architecture with the Bidirectional Encoder Representations from Transformers (BERT) layer in order to better utilize the temporal information with BERT's attention mechanism.
Posted Content
Temporal Convolution Based Action Proposal: Submission to ActivityNet 2017.
Tianwei Lin,Xu Zhao,Zheng Shou +2 more
TL;DR: The approach achieves the state-of-the-art performances on both temporal action proposal task and temporal action localization task and proposes a new proposal model based on temporal convolutional network.
References
More filters
Proceedings Article
ImageNet Classification with Deep Convolutional Neural Networks
TL;DR: The state-of-the-art performance of CNNs was achieved by Deep Convolutional Neural Networks (DCNNs) as discussed by the authors, which consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax.
Proceedings Article
Very Deep Convolutional Networks for Large-Scale Image Recognition
Karen Simonyan,Andrew Zisserman +1 more
TL;DR: This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.
Proceedings Article
Very Deep Convolutional Networks for Large-Scale Image Recognition
Karen Simonyan,Andrew Zisserman +1 more
TL;DR: In this paper, the authors investigated the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting and showed that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 layers.
Proceedings ArticleDOI
Going deeper with convolutions
Christian Szegedy,Wei Liu,Yangqing Jia,Pierre Sermanet,Scott Reed,Dragomir Anguelov,Dumitru Erhan,Vincent Vanhoucke,Andrew Rabinovich +8 more
TL;DR: Inception as mentioned in this paper is a deep convolutional neural network architecture that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14).
Proceedings Article
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Sergey Ioffe,Christian Szegedy +1 more
TL;DR: Applied to a state-of-the-art image classification model, Batch Normalization achieves the same accuracy with 14 times fewer training steps, and beats the original model by a significant margin.