Filters








262 Hits in 3.6 sec

Semantic Segmentation of RGBD Videos with Recurrent Fully Convolutional Neural Networks

Ekrem Emre Yurdakul, Yucel Yemez
2017 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)  
In this work, we use a synthetic RGBD video dataset to investigate the contribution of depth and temporal information to the video segmentation task using convolutional and recurrent neural network architectures  ...  Semantic segmentation of videos using neural networks is currently a popular task, the work done in this field is however mostly on RGB videos.  ...  In this work, we address the problem of pixel-level semantic segmentation on RGBD videos using both fully convolutional and recurrent fully convolutional neural networks.  ... 
doi:10.1109/iccvw.2017.51 dblp:conf/iccvw/YurdakulY17 fatcat:agss7hzsavh5vjvdfx7z7rl5x4

STFCN: Spatio-Temporal FCN for Semantic Video Segmentation [article]

Mohsen Fayyaz, Mohammad Hajizadeh Saffar, Mohammad Sabokrou, Mahmood Fathy, Reinhard Klette, Fay Huang
2016 arXiv   pre-print
We propose a module based on a long short-term memory (LSTM) architecture of a recurrent neural network for interpreting the temporal characteristics of video frames over time.  ...  Our key insight is to build spatio-temporal convolutional networks (spatio-temporal CNNs) that have an end-to-end architecture for semantic video segmentation.  ...  Neural networks are a very popular method for image segmentation, especially with the recent success of using convolutional neural network in the semantic segmentation field.  ... 
arXiv:1608.05971v2 fatcat:a4e734yj7za5rjw4edpxcra4ji

STD2P: RGBD Semantic Segmentation Using Spatio-Temporal Data-Driven Pooling

Yang He, Wei-Chen Chiu, Margret Keuper, Mario Fritz
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We propose a novel superpixel-based multi-view convolutional neural network for semantic image segmentation.  ...  The proposed network produces a high quality segmentation of a single image by leveraging information from additional views of the same scene.  ...  Context modeling for fully convolutional networks Fully convolutional networks (FCN) [26] , built on deep classification networks [19, 34] , carried their success forward to semantic segmentation networks  ... 
doi:10.1109/cvpr.2017.757 dblp:conf/cvpr/HeCKF17 fatcat:xxskxr3sfvcezky4ibus5vclm4

STD2P: RGBD Semantic Segmentation Using Spatio-Temporal Data-Driven Pooling [article]

Yang He, Wei-Chen Chiu, Margret Keuper, Mario Fritz
2017 arXiv   pre-print
We propose a novel superpixel-based multi-view convolutional neural network for semantic image segmentation.  ...  The proposed network produces a high quality segmentation of a single image by leveraging information from additional views of the same scene.  ...  Context modeling for fully convolutional networks Fully convolutional networks (FCN) [28] , built on deep classification networks [21, 36] , carried their success forward to semantic segmentation networks  ... 
arXiv:1604.02388v3 fatcat:rg76ankgrbbqzn6xa7uohruoeu

Survey on Semantic Segmentation using Deep Learning Techniques

Fahad Lateef, Yassine Ruichek
2019 Neurocomputing  
Semantic segmentation is a challenging task in computer vision systems.  ...  Finally, we conclude by discussing some of the open problems and their possible solutions.  ...  of work.  ... 
doi:10.1016/j.neucom.2019.02.003 fatcat:aelsfl7unvdw5j2rtyqhtgqrsm

Recurrent Scene Parsing with Perspective Understanding in the Loop

Shu Kong, Charless Fowlkes
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
We integrate this depth-aware gating into a recurrent convolutional neural network to perform semantic segmentation.  ...  Our recurrent module iteratively refines the segmentation results, leveraging the depth and semantic predictions from the previous iterations.  ...  The seminal DeepLab [6] model modifies the very deep residual neural network [16] for semantic segmentation using dilated or atrous convolution operators to maintain spatial resolution in high-level  ... 
doi:10.1109/cvpr.2018.00106 dblp:conf/cvpr/KongF18 fatcat:fnsxtabhvvg7fcnfwwadw5jd5a

Recurrent Scene Parsing with Perspective Understanding in the Loop [article]

Shu Kong, Charless Fowlkes
2017 arXiv   pre-print
We integrate this depth-aware gating into a recurrent convolutional neural network to perform semantic segmentation.  ...  Our recurrent module iteratively refines the segmentation results, leveraging the depth and semantic predictions from the previous iterations.  ...  The seminal DeepLab [6] model modifies the very deep residual neural network [16] for semantic segmentation using dilated or atrous convolution operators to maintain spatial resolution in high-level  ... 
arXiv:1705.07238v2 fatcat:h27ssaersrf2tdggyh6lwkzura

DA-RNN: Semantic Mapping with Data Associated Recurrent Neural Networks [article]

Yu Xiang, Dieter Fox
2017 arXiv   pre-print
DA-RNNs use a new recurrent neural network architecture for semantic labeling on RGB-D videos.  ...  In this work, we introduce Data Associated Recurrent Neural Networks (DA-RNNs), a novel framework for joint 3D scene mapping and semantic labeling.  ...  Single Frame Labeling with Fully Convolutional Networks The basis of our semantic labeling framework is a Fully Convolutional Network (FCN) for single frame labeling.  ... 
arXiv:1703.03098v2 fatcat:ilfisffvlbesph26mtzsunkthi

A real-time indoor scene analysis method based on RGBD stream

Chen Wang, Yue Qi
2019 IEEE Access  
It is noteworthy that our method can meet real-time requirements with frame rates of ≈25 Hz. INDEX TERMS Indoor scene analysis, camera tracking, recurrent neural networks, semantic segmentation.  ...  Then, we propose a structural constraint recurrent neural network (SC-RNN) to generate a semantic map for each frame.  ...  With the success of deep learning networks, especially convolutional neural networks (CNN), in the field of computer vision [31] - [33] , semantic segmentation has made new progress.  ... 
doi:10.1109/access.2019.2944140 fatcat:gqiptgzv6ffavd2lfrdx34wgfq

DA-RNN: Semantic Mapping with Data Associated Recurrent Neural Networks

Yu Xiang, Dieter Fox
2017 Robotics: Science and Systems XIII  
DA-RNNs use a new recurrent neural network architecture for semantic labeling on RGB-D videos.  ...  In this work, we introduce Data Associated Recurrent Neural Networks (DA-RNNs), a novel framework for joint 3D scene mapping and semantic labeling.  ...  Single Frame Labeling with Fully Convolutional Networks The basis of our semantic labeling framework is a Fully Convolutional Network (FCN) for single frame labeling.  ... 
doi:10.15607/rss.2017.xiii.013 dblp:conf/rss/XiangF17 fatcat:5jcxse4pqndr5ln26u3w6yxodm

Action Recognition by an Attention-Aware Temporal Weighted Convolutional Neural Network

Le Wang, Jinliang Zang, Qilin Zhang, Zhenxing Niu, Gang Hua, Nanning Zheng
2018 Sensors  
Research in human action recognition has accelerated significantly since the introduction of powerful machine learning tools such as Convolutional Neural Networks (CNNs).  ...  on more relevant video segments.  ...  Likewise, Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) have been popular in the video classification and detection task [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16]  ... 
doi:10.3390/s18071979 pmid:29933555 pmcid:PMC6069475 fatcat:byyotu7o75amzpbtifmpkpyunm

DeepArSLR: A Novel Signer-Independent Deep Learning Framework for Isolated Arabic Sign Language Gestures Recognition

Saleh Aly, Walaa Aly
2020 IEEE Access  
and deep recurrent neural network.  ...  Extracting hand shape features is achieved using a single layer Convolutional Self-Organizing Map (CSOM) instead of relying on transfer learning of pretrained deep convolutional neural networks.  ...  ACKNOWLEDGMENT The authors extend their appreciation to the Deanship of Scientific Research at Majmaah University for funding this work.  ... 
doi:10.1109/access.2020.2990699 fatcat:rnov5ewiprdcrh2jf5h6em626q

Image Segmentation Using Deep Learning: A Survey [article]

Shervin Minaee, Yuri Boykov, Fatih Porikli, Antonio Plaza, Nasser Kehtarnavaz, Demetri Terzopoulos
2020 arXiv   pre-print
In this survey, we provide a comprehensive review of the literature at the time of this writing, covering a broad spectrum of pioneering works for semantic and instance-level segmentation, including fully  ...  convolutional pixel-labeling networks, encoder-decoder architectures, multi-scale and pyramid based approaches, recurrent networks, visual attention models, and generative models in adversarial settings  ...  DA-RNNs use a new recurrent neural network architecture for semantic labeling on RGB-D videos.  ... 
arXiv:2001.05566v5 fatcat:wiep26nijncwxjojxbzrqoonti

Attention-based Temporal Weighted Convolutional Neural Network for Action Recognition [article]

Jinliang Zang, Le Wang, Ziyi Liu, Qilin Zhang, Zhenxing Niu, Gang Hua, Nanning Zheng
2018 arXiv   pre-print
Research in human action recognition has accelerated significantly since the introduction of powerful machine learning tools such as Convolutional Neural Networks (CNNs).  ...  Our experiments show that the proposed attention mechanism contributes substantially to the performance gains with the more discriminative snippets by focusing on more relevant video segments.  ...  Acknowledgment This work was supported partly by NSFC Grants 61629301, 61773312, 91748208 and 61503296, China Postdoctoral Science Foundation Grant 2017T100752, and key project of Shaanxi provinceS2018  ... 
arXiv:1803.07179v1 fatcat:bpiocesvb5bsvfqwkanqolr2ry

Deep Learning based Monocular Depth Prediction: Datasets, Methods and Applications [article]

Qing Li, Jiasong Zhu, Jun Liu, Rui Cao, Qingquan Li, Sen Jia, Guoping Qiu
2020 arXiv   pre-print
They surpass traditional machine learning-based methods by a large margin in terms of accuracy and speed.  ...  Recently, monocular depth estimation has obtained great progress owing to the rapid development of deep learning techniques.  ...  The fine-scale network refines the coarse depth prediction result with local information through a fully convolutional neural network.  ... 
arXiv:2011.04123v1 fatcat:by6swdegvvdrxk73ti46k2rj2e
« Previous Showing results 1 — 15 out of 262 results