Deep Temporal Models (Benchmarks and Applica6ons Analysis)

Size: px
Start display at page:

Download "Deep Temporal Models (Benchmarks and Applica6ons Analysis)"

Transcription

1 Deep Temporal Models (Benchmarks and Applica6ons Analysis) Sek Chai SRI Interna6onal Presented at: NICE 2016, March 7, SRI International

2 Project Summary Goals Analyze Deep Temporal Models (DTMs). Find approaches to reduce training 6me, lower memory size, and use low precision. High dimensional Data Set Audio, video, gesture Benchmark Deep Temporal Models Analysis Benchmarks and Applica4ons Analysis Processor Architecture RNN, LTSM, CRBM SRI Interna4onal Sek Chai (PI) Mohamed Amer David Zhang Tim Shields U. Guelph Graham Taylor Dhanesh Ramachandram U. Montreal Roland Memisevic Yoshua Bengio SRI International

3 Seeing Humans ChaLearn This dataset consists of a single user is recorded in front of a depth camera, performing natural communica6ve gestures and speaking in fluent Italian. The dataset focuses on the user independent automa6c recogni6on of a vocabulary of 20 Italian cultural/anthropological signs in image sequences. Challenges: Mul6modal visual cues (RGBD) and audio Mul6-6mescale, unreliable depth cues No informa6on about the number of gestures within each sequence High intra-class variability of gesture samples Low inter-class variability for some gesture categories. Several distractor gestures (out of the vocabulary) are present. Image: Neverova et al. (2015) S. Escalera, et al., "ChaLearn Looking at People Challenge 2014: Dataset and Results", ECCV-W SRI International

4 DeepGesture Architecture (for ChaLearn Dataset) temporal strides Valida6on Error, % N. Neverova, et al. (2015), ModDrop: Adap6ve Mul6modal Gesture Recogni6on, IEEE PAMI (In Press) State-of-Art 88.1% recogni4on rate 4 Training Stage Key Insights We adopted a strategy where like modali6es are fused first; it resembles brain s mul6- modal fusion strategy. Most previous work on mul6-model learning has fused data At the input feature level (early fusion); or At the level of per-modality classifier outputs (late fusion) 2016 SRI International

5 Example Training Complexity (for ChaLearn Dataset) Classifier Description Modality Training Data Size (GB) Training Time Epochs Time(sec)/ epoch Motion detect Skeleton Video Feature Video Shallow MLP used to detect the startframe Motion Capture GB 20,174 sec and stop-frame of a given gesture (5.6hrs) in an action set. Convolutional network which is trained to extract features from motion capture data (Path M) 3D convolutional layer followed by 2D convolution layer which uses depth and intensity video (ConvC1 -> ConvC2, ConD1- >ConvD2 ) Shared hidden layers which uses inputs from previous convolutional layers (HLV1 + HLV2) Multimodal Corresponds to the fully connected shared hidden layer where multimodal inputs are fused. (HLS) Motion Capture GB 69,344 sec (19.25hrs) intensity+depth video Intensity+depth video GB 82,655 sec (22.9 hrs) GB 170,223 sec (47.28 hrs) all GB 93,247 (25.9 hrs) 200, , Summary Total 5 days to process 42GB training data on Sharcnet Copper Cluster (064 GPUs, 128 CPU cores, 24 cores/node, 64 GB/node, x86, 080 TB RAID Ahached Storage, InfiniBand, 4 Tesla K80s/node). Total # Parameters = 7,836, SRI International

6 Low Precision Neural Networks Needs: Memory is main bohleneck, especially for embedded solu6ons. Es6mates 1B connec6on neural network consumes 12W*. Current Approaches Stochas6c rounding Gupta, et al. (arxiv 2015) Network Pruning *Image: Han, et al. (NIPS 2015) Binary Connect Image: Courbariaux, et al. (NIPS 2015) 2016 SRI International 6

7 Subband /Wavelet Decomposi6on Subband decomposi6on enables data reduc6on by discarding informa6on about certain frequencies where human visual system is less sensi6ve.* Can we do the same for learnt representa6ons? * Good representa6on : Used in Image Compression[1], Reconstruc6on[2] and Fusion[3] extensively. [1] "The Laplacian Pyramid as a Compact Image Code", Burt et. al. [2] Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks, Denton et. al. [3] Image fusion: algorithms and applications, Stathaki SRI International 7

8 Conven6onal Approach for MNIST data I 0 *Image: Lecun, et al. (1998) The MNIST database contains black and white handwrihen digits, normalized to 20x20 pixel size. There are 60,000 training images and 10,000 tes6ng images SRI International 8

9 Separate Networks for Subband Learning Our Approach Image (I 0 ) Laplacian (L 0 ) Gaussian (G 1 ) 28x28 32x x16 Captures Edges Spa6al Informa6on L o G 1 Background Cues + Fusion Basic Idea: We separate imagery into different frequency bands (e.g. with different informa6on content) such that the neural net can be.er learn using less bits SRI International 9

10 Valida6on Error vs Epochs in MNIST I 0 L 0 G 1 + Valida4on Error (Log scale) 28x28 32x32 16x16 Discussions CNN trained on Laplacian focusses more on edges (good feature). CNN(L 0 ) beats CNN(I 0 ) on this dataset. --- Gaussian G 1 -o- Original I 0 -x- Laplace L 0 -x- Fusion Epochs 2016 SRI International 10

11 Robustness to Low Bit Precision Weights Stochas4c Rounding azer final epoch Weight bits 32bit 16bit 8bit 4bit Original GBlur Laplace Fusion Stochas4c Rounding azer every epoch Weight bits 32bit 16bit 8bit 4bit Original GBlur Laplace Fusion Discussions Fusion results are comparable to original, using half the number of bits. Stochas6c rounding aver every epoch guides the learning, and is especially useful for low precision. Simple Fusion : Equi-Weighted average of sovmax output Scores 0.5*(s 1 (x)+s 2 (y)), s 1 (x),s 2 (x) in [0,1] 10 and s(x) 1 = SRI International 11

12 Cifar-10 data set The CIFAR-10 dataset consists of x32 color images in 10 classes, with 6000 images per class. There are training images and test images. Discussions: Cluhered color images, and more challenging than MNIST. Contains background cues and context that can help recogni6on, (e.g. blue sky for airplane or water for ship). Architecture is the same as before (LeNet-5)with 30 and 60 feature maps. Our goal is to show compara6ve results for low precision. There are no data augmenta6on SRI International 12

13 CIFAR-10 Performance Gaussian blur removes noise in clu.er Test error Laplacian enhances foreground Clu.ered image combined with high learning rate Epoch Original Laplacian Gblur Fusion 2016 SRI International 13

14 Conclusion Hybrid-mul6modal neural networks improves algorithmic performance. Fusion of learnt representa6ons is important. Low precision networks shows promise. Stop by and visit the poster at NICE Chai, et al., "Low Precision Neural Networks using Subband DecomposiHon", CogArch, April SRI International 14

LEARNING DEEP MULTI-MODAL FUSION ARCHITECTURES

LEARNING DEEP MULTI-MODAL FUSION ARCHITECTURES LEARNING DEEP MULTI-MODAL FUSION ARCHITECTURES GRAHAM TAYLOR SCHOOL OF ENGINEERING UNIVERSITY OF GUELPH Joint work with: Natalia Neverova (Facebook AI Research), Christian Wolf (INSA-Lyon) Dhanesh Ramachandram

More information

Structure Optimization for Deep Multimodal Fusion Networks using Graph-Induced Kernels

Structure Optimization for Deep Multimodal Fusion Networks using Graph-Induced Kernels Structure Optimization for Deep Multimodal Fusion Networks using Graph-Induced Kernels Dhanesh Ramachandram 1, Michal Lisicki 1, Timothy J. Shields, Mohamed R. Amer and Graham W. Taylor1 1- Machine Learning

More information

RGBD Occlusion Detection via Deep Convolutional Neural Networks

RGBD Occlusion Detection via Deep Convolutional Neural Networks 1 RGBD Occlusion Detection via Deep Convolutional Neural Networks Soumik Sarkar 1,2, Vivek Venugopalan 1, Kishore Reddy 1, Michael Giering 1, Julian Ryde 3, Navdeep Jaitly 4,5 1 United Technologies Research

More information

Large-scale gesture recognition based on Multimodal data with C3D and TSN

Large-scale gesture recognition based on Multimodal data with C3D and TSN Large-scale gesture recognition based on Multimodal data with C3D and TSN July 6, 2017 1 Team details Team name ASU Team leader name Yunan Li Team leader address, phone number and email address: Xidian

More information

Energy- Aware Time Change Detec4on Using Synthe4c Aperture Radar On High- Performance Heterogeneous Architectures: A DDDAS Approach

Energy- Aware Time Change Detec4on Using Synthe4c Aperture Radar On High- Performance Heterogeneous Architectures: A DDDAS Approach Energy- Aware Time Change Detec4on Using Synthe4c Aperture Radar On High- Performance Heterogeneous Architectures: A DDDAS Approach Sanjay Ranka (PI) Sartaj Sahni (Co- PI) Mark Schmalz (Co- PI) University

More information

Deep Tracking: Biologically Inspired Tracking with Deep Convolutional Networks

Deep Tracking: Biologically Inspired Tracking with Deep Convolutional Networks Deep Tracking: Biologically Inspired Tracking with Deep Convolutional Networks Si Chen The George Washington University sichen@gwmail.gwu.edu Meera Hahn Emory University mhahn7@emory.edu Mentor: Afshin

More information

Convolutional-Recursive Deep Learning for 3D Object Classification

Convolutional-Recursive Deep Learning for 3D Object Classification Convolutional-Recursive Deep Learning for 3D Object Classification Richard Socher, Brody Huval, Bharath Bhat, Christopher D. Manning, Andrew Y. Ng NIPS 2012 Iro Armeni, Manik Dhar Motivation Hand-designed

More information

Accelerating Binarized Convolutional Neural Networks with Software-Programmable FPGAs

Accelerating Binarized Convolutional Neural Networks with Software-Programmable FPGAs Accelerating Binarized Convolutional Neural Networks with Software-Programmable FPGAs Ritchie Zhao 1, Weinan Song 2, Wentao Zhang 2, Tianwei Xing 3, Jeng-Hau Lin 4, Mani Srivastava 3, Rajesh Gupta 4, Zhiru

More information

Convolutional Neural Networks

Convolutional Neural Networks Lecturer: Barnabas Poczos Introduction to Machine Learning (Lecture Notes) Convolutional Neural Networks Disclaimer: These notes have not been subjected to the usual scrutiny reserved for formal publications.

More information

Video Object Segmentation using Deep Learning

Video Object Segmentation using Deep Learning Video Object Segmentation using Deep Learning Update Presentation, Week 2 Zack While Advised by: Rui Hou, Dr. Chen Chen, and Dr. Mubarak Shah May 26, 2017 Youngstown State University 1 Table of Contents

More information

A Deep Learning Framework for Authorship Classification of Paintings

A Deep Learning Framework for Authorship Classification of Paintings A Deep Learning Framework for Authorship Classification of Paintings Kai-Lung Hua ( 花凱龍 ) Dept. of Computer Science and Information Engineering National Taiwan University of Science and Technology Taipei,

More information

Machine learning for image- based localiza4on. Juho Kannala May 15, 2017

Machine learning for image- based localiza4on. Juho Kannala May 15, 2017 Machine learning for image- based localiza4on Juho Kannala May 15, 2017 Contents Problem sebng (What?) Mo4va4on & applica4ons (Why?) Previous work & background (How?) Our own studies and results Open ques4ons

More information

Outline GF-RNN ReNet. Outline

Outline GF-RNN ReNet. Outline Outline Gated Feedback Recurrent Neural Networks. arxiv1502. Introduction: RNN & Gated RNN Gated Feedback Recurrent Neural Networks (GF-RNN) Experiments: Character-level Language Modeling & Python Program

More information

Computer Vision Lecture 16

Computer Vision Lecture 16 Computer Vision Lecture 16 Deep Learning for Object Categorization 14.01.2016 Bastian Leibe RWTH Aachen http://www.vision.rwth-aachen.de leibe@vision.rwth-aachen.de Announcements Seminar registration period

More information

Video Gesture Recognition with RGB-D-S Data Based on 3D Convolutional Networks

Video Gesture Recognition with RGB-D-S Data Based on 3D Convolutional Networks Video Gesture Recognition with RGB-D-S Data Based on 3D Convolutional Networks August 16, 2016 1 Team details Team name FLiXT Team leader name Yunan Li Team leader address, phone number and email address:

More information

OPTIMIZED GPU KERNELS FOR DEEP LEARNING. Amir Khosrowshahi

OPTIMIZED GPU KERNELS FOR DEEP LEARNING. Amir Khosrowshahi OPTIMIZED GPU KERNELS FOR DEEP LEARNING Amir Khosrowshahi GTC 17 Mar 2015 Outline About nervana Optimizing deep learning at assembler level Limited precision for deep learning neon benchmarks 2 About nervana

More information

Fully Convolutional Networks for Semantic Segmentation

Fully Convolutional Networks for Semantic Segmentation Fully Convolutional Networks for Semantic Segmentation Jonathan Long* Evan Shelhamer* Trevor Darrell UC Berkeley Chaim Ginzburg for Deep Learning seminar 1 Semantic Segmentation Define a pixel-wise labeling

More information

Multilayer and Multimodal Fusion of Deep Neural Networks for Video Classification

Multilayer and Multimodal Fusion of Deep Neural Networks for Video Classification Multilayer and Multimodal Fusion of Deep Neural Networks for Video Classification Xiaodong Yang, Pavlo Molchanov, Jan Kautz INTELLIGENT VIDEO ANALYTICS Surveillance event detection Human-computer interaction

More information

Multi-scale deep learning for gesture detection and localization

Multi-scale deep learning for gesture detection and localization Multi-scale deep learning for gesture detection and localization 1,2 Natalia Neverova 1,2 Christian Wolf 3 Graham W. Taylor 4 Florian Nebout 1 Université de Lyon, CNRS, France firstname.surname@liris.cnrs.fr

More information

CS231N Section. Video Understanding 6/1/2018

CS231N Section. Video Understanding 6/1/2018 CS231N Section Video Understanding 6/1/2018 Outline Background / Motivation / History Video Datasets Models Pre-deep learning CNN + RNN 3D convolution Two-stream What we ve seen in class so far... Image

More information

Shape Context Matching For Efficient OCR

Shape Context Matching For Efficient OCR Matching For Efficient OCR May 14, 2012 Matching For Efficient OCR Table of contents 1 Motivation Background 2 What is a? Matching s Simliarity Measure 3 Matching s via Pyramid Matching Matching For Efficient

More information

arxiv: v1 [cs.cv] 19 Jun 2018

arxiv: v1 [cs.cv] 19 Jun 2018 Multimodal feature fusion for CNN-based gait recognition: an empirical comparison F.M. Castro a,, M.J. Marín-Jiménez b, N. Guil a, N. Pérez de la Blanca c a Department of Computer Architecture, University

More information

Ryerson University CP8208. Soft Computing and Machine Intelligence. Naive Road-Detection using CNNS. Authors: Sarah Asiri - Domenic Curro

Ryerson University CP8208. Soft Computing and Machine Intelligence. Naive Road-Detection using CNNS. Authors: Sarah Asiri - Domenic Curro Ryerson University CP8208 Soft Computing and Machine Intelligence Naive Road-Detection using CNNS Authors: Sarah Asiri - Domenic Curro April 24 2016 Contents 1 Abstract 2 2 Introduction 2 3 Motivation

More information

Multimodal Gesture Recognition using Multi-stream Recurrent Neural Network

Multimodal Gesture Recognition using Multi-stream Recurrent Neural Network Multimodal Gesture Recognition using Multi-stream Recurrent Neural Network Noriki Nishida and Hideki Nakayama Machine Perception Group Graduate School of Information Science and Technology The University

More information

arxiv: v1 [stat.ml] 21 Feb 2018

arxiv: v1 [stat.ml] 21 Feb 2018 Detecting Learning vs Memorization in Deep Neural Networks using Shared Structure Validation Sets arxiv:2.0774v [stat.ml] 2 Feb 8 Elias Chaibub Neto e-mail: elias.chaibub.neto@sagebase.org, Sage Bionetworks

More information

THE MNIST DATABASE of handwritten digits Yann LeCun, Courant Institute, NYU Corinna Cortes, Google Labs, New York

THE MNIST DATABASE of handwritten digits Yann LeCun, Courant Institute, NYU Corinna Cortes, Google Labs, New York THE MNIST DATABASE of handwritten digits Yann LeCun, Courant Institute, NYU Corinna Cortes, Google Labs, New York The MNIST database of handwritten digits, available from this page, has a training set

More information

Machine Learning 13. week

Machine Learning 13. week Machine Learning 13. week Deep Learning Convolutional Neural Network Recurrent Neural Network 1 Why Deep Learning is so Popular? 1. Increase in the amount of data Thanks to the Internet, huge amount of

More information

Deep Neural Networks:

Deep Neural Networks: Deep Neural Networks: Part II Convolutional Neural Network (CNN) Yuan-Kai Wang, 2016 Web site of this course: http://pattern-recognition.weebly.com source: CNN for ImageClassification, by S. Lazebnik,

More information

Xiaowei Hu* Lei Zhu* Chi-Wing Fu Jing Qin Pheng-Ann Heng

Xiaowei Hu* Lei Zhu* Chi-Wing Fu Jing Qin Pheng-Ann Heng Direction-aware Spatial Context Features for Shadow Detection Xiaowei Hu* Lei Zhu* Chi-Wing Fu Jing Qin Pheng-Ann Heng The Chinese University of Hong Kong The Hong Kong Polytechnic University Shenzhen

More information

Low-Power Neural Processor for Embedded Human and Face detection

Low-Power Neural Processor for Embedded Human and Face detection Low-Power Neural Processor for Embedded Human and Face detection Olivier Brousse 1, Olivier Boisard 1, Michel Paindavoine 1,2, Jean-Marc Philippe, Alexandre Carbon (1) GlobalSensing Technologies (GST)

More information

Convolutional Neural Networks. Computer Vision Jia-Bin Huang, Virginia Tech

Convolutional Neural Networks. Computer Vision Jia-Bin Huang, Virginia Tech Convolutional Neural Networks Computer Vision Jia-Bin Huang, Virginia Tech Today s class Overview Convolutional Neural Network (CNN) Training CNN Understanding and Visualizing CNN Image Categorization:

More information

Profiling the Performance of Binarized Neural Networks. Daniel Lerner, Jared Pierce, Blake Wetherton, Jialiang Zhang

Profiling the Performance of Binarized Neural Networks. Daniel Lerner, Jared Pierce, Blake Wetherton, Jialiang Zhang Profiling the Performance of Binarized Neural Networks Daniel Lerner, Jared Pierce, Blake Wetherton, Jialiang Zhang 1 Outline Project Significance Prior Work Research Objectives Hypotheses Testing Framework

More information

CNN for Low Level Image Processing. Huanjing Yue

CNN for Low Level Image Processing. Huanjing Yue CNN for Low Level Image Processing Huanjing Yue 2017.11 1 Deep Learning for Image Restoration General formulation: min Θ L( x, x) s. t. x = F(y; Θ) Loss function Parameters to be learned Key issues The

More information

Visible and Long-Wave Infrared Image Fusion Schemes for Situational. Awareness

Visible and Long-Wave Infrared Image Fusion Schemes for Situational. Awareness Visible and Long-Wave Infrared Image Fusion Schemes for Situational Awareness Multi-Dimensional Digital Signal Processing Literature Survey Nathaniel Walker The University of Texas at Austin nathaniel.walker@baesystems.com

More information

Implementation of Image Fusion Algorithm Using Laplace Transform

Implementation of Image Fusion Algorithm Using Laplace Transform Implementation of Image Fusion Algorithm Using Laplace Transform S.Santhosh kumar M.Tech, Srichaitanya Institute of Technology & Sciences, Karimnagar,Telangana. G Sahithi, M.Tech Assistant Professor, Srichaitanya

More information

Deep Learning For Video Classification. Presented by Natalie Carlebach & Gil Sharon

Deep Learning For Video Classification. Presented by Natalie Carlebach & Gil Sharon Deep Learning For Video Classification Presented by Natalie Carlebach & Gil Sharon Overview Of Presentation Motivation Challenges of video classification Common datasets 4 different methods presented in

More information

An Empirical Evaluation of Deep Architectures on Problems with Many Factors of Variation

An Empirical Evaluation of Deep Architectures on Problems with Many Factors of Variation An Empirical Evaluation of Deep Architectures on Problems with Many Factors of Variation Hugo Larochelle, Dumitru Erhan, Aaron Courville, James Bergstra, and Yoshua Bengio Université de Montréal 13/06/2007

More information

ImageNet Classification with Deep Convolutional Neural Networks

ImageNet Classification with Deep Convolutional Neural Networks ImageNet Classification with Deep Convolutional Neural Networks Alex Krizhevsky Ilya Sutskever Geoffrey Hinton University of Toronto Canada Paper with same name to appear in NIPS 2012 Main idea Architecture

More information

Generative Adversarial Text to Image Synthesis

Generative Adversarial Text to Image Synthesis Generative Adversarial Text to Image Synthesis Scott Reed, Zeynep Akata, Xinchen Yan, Lajanugen Logeswaran, Bernt Schiele, Honglak Lee Presented by: Jingyao Zhan Contents Introduction Related Work Method

More information

Advanced Introduction to Machine Learning, CMU-10715

Advanced Introduction to Machine Learning, CMU-10715 Advanced Introduction to Machine Learning, CMU-10715 Deep Learning Barnabás Póczos, Sept 17 Credits Many of the pictures, results, and other materials are taken from: Ruslan Salakhutdinov Joshua Bengio

More information

Exploring Capsules. Binghui Peng Runzhou Tao Shunyu Yao IIIS, Tsinghua University {pbh15, trz15,

Exploring Capsules. Binghui Peng Runzhou Tao Shunyu Yao IIIS, Tsinghua University {pbh15, trz15, Exploring Capsules Binghui Peng Runzhou Tao Shunyu Yao IIIS, Tsinghua University {pbh15, trz15, yao-sy15}@mails.tsinghua.edu.cn 1 Introduction Nowadays, convolutional neural networks (CNNs) have received

More information

Single Object Tracking with Organic Optic Attenuation

Single Object Tracking with Organic Optic Attenuation Single Object Tracking with Organic Optic Attenuation Note: DEMO GIFS Have been removed due to making the presentation too large to upload to blackboard! (other gifs have been lossy-compressed) Ibraheem

More information

LSTM: An Image Classification Model Based on Fashion-MNIST Dataset

LSTM: An Image Classification Model Based on Fashion-MNIST Dataset LSTM: An Image Classification Model Based on Fashion-MNIST Dataset Kexin Zhang, Research School of Computer Science, Australian National University Kexin Zhang, U6342657@anu.edu.au Abstract. The application

More information

ITSME: Mul*modal and Unobtrusive Smartphone User Authen*ca*on

ITSME: Mul*modal and Unobtrusive Smartphone User Authen*ca*on ITSME: Mul*modal and Unobtrusive Smartphone User Authen*ca*on A

More information

ChaLearn Looking at People Workshop

ChaLearn Looking at People Workshop ChaLearn Looking at People Workshop Cultural Event Recognition (Demo) Junior Fabian, CVC, jfabian@cvc.uab.es, Hugo Escalante, INAOE, Xavier Baró, UOC, Sergio Escalera, CVC/UB, Jordi González, CVC, Pablo

More information

Deep Neural Network Hyperparameter Optimization with Genetic Algorithms

Deep Neural Network Hyperparameter Optimization with Genetic Algorithms Deep Neural Network Hyperparameter Optimization with Genetic Algorithms EvoDevo A Genetic Algorithm Framework Aaron Vose, Jacob Balma, Geert Wenes, and Rangan Sukumar Cray Inc. October 2017 Presenter Vose,

More information

Study of Residual Networks for Image Recognition

Study of Residual Networks for Image Recognition Study of Residual Networks for Image Recognition Mohammad Sadegh Ebrahimi Stanford University sadegh@stanford.edu Hossein Karkeh Abadi Stanford University hosseink@stanford.edu Abstract Deep neural networks

More information

Deep Learning on Graphs

Deep Learning on Graphs Deep Learning on Graphs with Graph Convolutional Networks Hidden layer Hidden layer Input Output ReLU ReLU, 22 March 2017 joint work with Max Welling (University of Amsterdam) BDL Workshop @ NIPS 2016

More information

Deep Learning With Noise

Deep Learning With Noise Deep Learning With Noise Yixin Luo Computer Science Department Carnegie Mellon University yixinluo@cs.cmu.edu Fan Yang Department of Mathematical Sciences Carnegie Mellon University fanyang1@andrew.cmu.edu

More information

Restricted Boltzmann Machines. Shallow vs. deep networks. Stacked RBMs. Boltzmann Machine learning: Unsupervised version

Restricted Boltzmann Machines. Shallow vs. deep networks. Stacked RBMs. Boltzmann Machine learning: Unsupervised version Shallow vs. deep networks Restricted Boltzmann Machines Shallow: one hidden layer Features can be learned more-or-less independently Arbitrary function approximator (with enough hidden units) Deep: two

More information

Hybrid Biometric Person Authentication Using Face and Voice Features

Hybrid Biometric Person Authentication Using Face and Voice Features Paper presented in the Third International Conference, Audio- and Video-Based Biometric Person Authentication AVBPA 2001, Halmstad, Sweden, proceedings pages 348-353, June 2001. Hybrid Biometric Person

More information

Convolutional Networks in Scene Labelling

Convolutional Networks in Scene Labelling Convolutional Networks in Scene Labelling Ashwin Paranjape Stanford ashwinpp@stanford.edu Ayesha Mudassir Stanford aysh@stanford.edu Abstract This project tries to address a well known problem of multi-class

More information

Capsule Networks. Eric Mintun

Capsule Networks. Eric Mintun Capsule Networks Eric Mintun Motivation An improvement* to regular Convolutional Neural Networks. Two goals: Replace max-pooling operation with something more intuitive. Keep more info about an activated

More information

Clustering algorithms and autoencoders for anomaly detection

Clustering algorithms and autoencoders for anomaly detection Clustering algorithms and autoencoders for anomaly detection Alessia Saggio Lunch Seminars and Journal Clubs Université catholique de Louvain, Belgium 3rd March 2017 a Outline Introduction Clustering algorithms

More information

Deep Learning. Deep Learning. Practical Application Automatically Adding Sounds To Silent Movies

Deep Learning. Deep Learning. Practical Application Automatically Adding Sounds To Silent Movies http://blog.csdn.net/zouxy09/article/details/8775360 Automatic Colorization of Black and White Images Automatically Adding Sounds To Silent Movies Traditionally this was done by hand with human effort

More information

6. Convolutional Neural Networks

6. Convolutional Neural Networks 6. Convolutional Neural Networks CS 519 Deep Learning, Winter 2017 Fuxin Li With materials from Zsolt Kira Quiz coming up Next Thursday (2/2) 20 minutes Topics: Optimization Basic neural networks No Convolutional

More information

Deep Learning for Visual Manipulation and Synthesis

Deep Learning for Visual Manipulation and Synthesis Deep Learning for Visual Manipulation and Synthesis Jun-Yan Zhu 朱俊彦 UC Berkeley 2017/01/11 @ VALSE What is visual manipulation? Image Editing Program input photo User Input result Desired output: stay

More information

Convolu'onal Neural Networks

Convolu'onal Neural Networks Convolu'onal Neural Networks Dr. Kira Radinsky CTO SalesPredict Visi8ng Professor/Scien8st Technion Slides were adapted from Fei-Fei Li & Andrej Karpathy & Jus8n Johnson A bit of history: Hubel & Wiesel,

More information

Deep Incremental Scene Understanding. Federico Tombari & Christian Rupprecht Technical University of Munich, Germany

Deep Incremental Scene Understanding. Federico Tombari & Christian Rupprecht Technical University of Munich, Germany Deep Incremental Scene Understanding Federico Tombari & Christian Rupprecht Technical University of Munich, Germany C. Couprie et al. "Toward Real-time Indoor Semantic Segmentation Using Depth Information"

More information

Neural Networks for Machine Learning. Lecture 5a Why object recogni:on is difficult. Geoffrey Hinton with Ni:sh Srivastava Kevin Swersky

Neural Networks for Machine Learning. Lecture 5a Why object recogni:on is difficult. Geoffrey Hinton with Ni:sh Srivastava Kevin Swersky Neural Networks for Machine Learning Lecture 5a Why object recogni:on is difficult Geoffrey Hinton with Ni:sh Srivastava Kevin Swersky Things that make it hard to recognize objects Segmenta:on: Real scenes

More information

Two-Stream Convolutional Networks for Action Recognition in Videos

Two-Stream Convolutional Networks for Action Recognition in Videos Two-Stream Convolutional Networks for Action Recognition in Videos Karen Simonyan Andrew Zisserman Cemil Zalluhoğlu Introduction Aim Extend deep Convolution Networks to action recognition in video. Motivation

More information

DEEP LEARNING REVIEW. Yann LeCun, Yoshua Bengio & Geoffrey Hinton Nature Presented by Divya Chitimalla

DEEP LEARNING REVIEW. Yann LeCun, Yoshua Bengio & Geoffrey Hinton Nature Presented by Divya Chitimalla DEEP LEARNING REVIEW Yann LeCun, Yoshua Bengio & Geoffrey Hinton Nature 2015 -Presented by Divya Chitimalla What is deep learning Deep learning allows computational models that are composed of multiple

More information

COMPARATIVE DEEP LEARNING FOR CONTENT- BASED MEDICAL IMAGE RETRIEVAL

COMPARATIVE DEEP LEARNING FOR CONTENT- BASED MEDICAL IMAGE RETRIEVAL 1 COMPARATIVE DEEP LEARNING FOR CONTENT- BASED MEDICAL IMAGE RETRIEVAL ADITYA SRIRAM DECEMBER 1 st, 2016 Aditya Sriram CS846 Software Engineering for Big Data December 1, 2016 TOPICS 2 Paper Synopsis Content-Based

More information

Constrained Convolutional Neural Networks for Weakly Supervised Segmentation. Deepak Pathak, Philipp Krähenbühl and Trevor Darrell

Constrained Convolutional Neural Networks for Weakly Supervised Segmentation. Deepak Pathak, Philipp Krähenbühl and Trevor Darrell Constrained Convolutional Neural Networks for Weakly Supervised Segmentation Deepak Pathak, Philipp Krähenbühl and Trevor Darrell 1 Multi-class Image Segmentation Assign a class label to each pixel in

More information

Adaptive Gesture Recognition System Integrating Multiple Inputs

Adaptive Gesture Recognition System Integrating Multiple Inputs Adaptive Gesture Recognition System Integrating Multiple Inputs Master Thesis - Colloquium Tobias Staron University of Hamburg Faculty of Mathematics, Informatics and Natural Sciences Technical Aspects

More information

Video- to- Video Face Matching: Establishing a Baseline for Unconstrained Face Recogni:on

Video- to- Video Face Matching: Establishing a Baseline for Unconstrained Face Recogni:on Video- to- Video Face Matching: Establishing a Baseline for Unconstrained Face Recogni:on Lacey Best- Rowden, Brendan Klare, Joshua Klontz, and Anil K. Jain Biometrics: Theory, Applica:ons, and Systems

More information

Geometric VLAD for Large Scale Image Search. Zixuan Wang 1, Wei Di 2, Anurag Bhardwaj 2, Vignesh Jagadesh 2, Robinson Piramuthu 2

Geometric VLAD for Large Scale Image Search. Zixuan Wang 1, Wei Di 2, Anurag Bhardwaj 2, Vignesh Jagadesh 2, Robinson Piramuthu 2 Geometric VLAD for Large Scale Image Search Zixuan Wang 1, Wei Di 2, Anurag Bhardwaj 2, Vignesh Jagadesh 2, Robinson Piramuthu 2 1 2 Our Goal 1) Robust to various imaging conditions 2) Small memory footprint

More information

arxiv: v1 [cs.cv] 26 Jun 2017

arxiv: v1 [cs.cv] 26 Jun 2017 Skeleton-Based Action Recognition Using Spatio-Temporal LSTM Network with Trust Gates Jun Liu, Amir Shahroudy, Dong Xu, Alex C. Kot, and Gang Wang arxiv:706.0876v [cs.cv] 6 Jun 07 Abstract Skeleton-based

More information

Deep Learning with Tensorflow AlexNet

Deep Learning with Tensorflow   AlexNet Machine Learning and Computer Vision Group Deep Learning with Tensorflow http://cvml.ist.ac.at/courses/dlwt_w17/ AlexNet Krizhevsky, Alex, Ilya Sutskever, and Geoffrey E. Hinton, "Imagenet classification

More information

Feature Visualization

Feature Visualization CreativeAI: Deep Learning for Graphics Feature Visualization Niloy Mitra Iasonas Kokkinos Paul Guerrero Nils Thuerey Tobias Ritschel UCL UCL UCL TU Munich UCL Timetable Theory and Basics State of the Art

More information

A Deep Learning primer

A Deep Learning primer A Deep Learning primer Riccardo Zanella r.zanella@cineca.it SuperComputing Applications and Innovation Department 1/21 Table of Contents Deep Learning: a review Representation Learning methods DL Applications

More information

Bidirectional Recurrent Convolutional Networks for Video Super-Resolution

Bidirectional Recurrent Convolutional Networks for Video Super-Resolution Bidirectional Recurrent Convolutional Networks for Video Super-Resolution Qi Zhang & Yan Huang Center for Research on Intelligent Perception and Computing (CRIPAC) National Laboratory of Pattern Recognition

More information

Facial Key Points Detection using Deep Convolutional Neural Network - NaimishNet

Facial Key Points Detection using Deep Convolutional Neural Network - NaimishNet 1 Facial Key Points Detection using Deep Convolutional Neural Network - NaimishNet Naimish Agarwal, IIIT-Allahabad (irm2013013@iiita.ac.in) Artus Krohn-Grimberghe, University of Paderborn (artus@aisbi.de)

More information

Deep Learning on SHARCNET:

Deep Learning on SHARCNET: Deep Learning on SHARCNET: Best Practices Fei Mao Outlines What does SHARCNET have? - Hardware/software resources now and future How to run a job? - A torch7 example How to train in parallel: - A Theano-based

More information

COMP9444 Neural Networks and Deep Learning 7. Image Processing. COMP9444 c Alan Blair, 2017

COMP9444 Neural Networks and Deep Learning 7. Image Processing. COMP9444 c Alan Blair, 2017 COMP9444 Neural Networks and Deep Learning 7. Image Processing COMP9444 17s2 Image Processing 1 Outline Image Datasets and Tasks Convolution in Detail AlexNet Weight Initialization Batch Normalization

More information

Deep Learning. Volker Tresp Summer 2014

Deep Learning. Volker Tresp Summer 2014 Deep Learning Volker Tresp Summer 2014 1 Neural Network Winter and Revival While Machine Learning was flourishing, there was a Neural Network winter (late 1990 s until late 2000 s) Around 2010 there

More information

Regionlet Object Detector with Hand-crafted and CNN Feature

Regionlet Object Detector with Hand-crafted and CNN Feature Regionlet Object Detector with Hand-crafted and CNN Feature Xiaoyu Wang Research Xiaoyu Wang Research Ming Yang Horizon Robotics Shenghuo Zhu Alibaba Group Yuanqing Lin Baidu Overview of this section Regionlet

More information

Learning visual odometry with a convolutional network

Learning visual odometry with a convolutional network Learning visual odometry with a convolutional network Kishore Konda 1, Roland Memisevic 2 1 Goethe University Frankfurt 2 University of Montreal konda.kishorereddy@gmail.com, roland.memisevic@gmail.com

More information

Detec%ng Wildlife in Uncontrolled Outdoor Video using Convolu%onal Neural Networks

Detec%ng Wildlife in Uncontrolled Outdoor Video using Convolu%onal Neural Networks Detec%ng Wildlife in Uncontrolled Outdoor Video using Convolu%onal Neural Networks Connor Bowley *, Alicia Andes +, Susan Ellis-Felege +, Travis Desell * Department of Computer Science * Department of

More information

GAN Related Works. CVPR 2018 & Selective Works in ICML and NIPS. Zhifei Zhang

GAN Related Works. CVPR 2018 & Selective Works in ICML and NIPS. Zhifei Zhang GAN Related Works CVPR 2018 & Selective Works in ICML and NIPS Zhifei Zhang Generative Adversarial Networks (GANs) 9/12/2018 2 Generative Adversarial Networks (GANs) Feedforward Backpropagation Real? z

More information

Progressive Generative Hashing for Image Retrieval

Progressive Generative Hashing for Image Retrieval Progressive Generative Hashing for Image Retrieval Yuqing Ma, Yue He, Fan Ding, Sheng Hu, Jun Li, Xianglong Liu 2018.7.16 01 BACKGROUND the NNS problem in big data 02 RELATED WORK Generative adversarial

More information

Depth Estimation from a Single Image Using a Deep Neural Network Milestone Report

Depth Estimation from a Single Image Using a Deep Neural Network Milestone Report Figure 1: The architecture of the convolutional network. Input: a single view image; Output: a depth map. 3 Related Work In [4] they used depth maps of indoor scenes produced by a Microsoft Kinect to successfully

More information

Decision Trees, Random Forests and Random Ferns. Peter Kovesi

Decision Trees, Random Forests and Random Ferns. Peter Kovesi Decision Trees, Random Forests and Random Ferns Peter Kovesi What do I want to do? Take an image. Iden9fy the dis9nct regions of stuff in the image. Mark the boundaries of these regions. Recognize and

More information

arxiv: v2 [cs.cv] 26 Apr 2018

arxiv: v2 [cs.cv] 26 Apr 2018 Motion Fused Frames: Data Level Fusion Strategy for Hand Gesture Recognition arxiv:1804.07187v2 [cs.cv] 26 Apr 2018 Okan Köpüklü Neslihan Köse Gerhard Rigoll Institute for Human-Machine Communication Technical

More information

Artificial Intelligence Introduction Handwriting Recognition Kadir Eren Unal ( ), Jakob Heyder ( )

Artificial Intelligence Introduction Handwriting Recognition Kadir Eren Unal ( ), Jakob Heyder ( ) Structure: 1. Introduction 2. Problem 3. Neural network approach a. Architecture b. Phases of CNN c. Results 4. HTM approach a. Architecture b. Setup c. Results 5. Conclusion 1.) Introduction Artificial

More information

GPU Accelerated Model Combination for Robust Speech Recognition and Keyword Search

GPU Accelerated Model Combination for Robust Speech Recognition and Keyword Search GPU Accelerated Model Combination for Robust Speech Recognition and Keyword Search Wonkyum Lee Jungsuk Kim Ian Lane Electrical and Computer Engineering Carnegie Mellon University March 26, 2014 @GTC2014

More information

Overcoming the Barriers of Graphs on GPUs: Delivering Graph Analy;cs 100X Faster and 40X Cheaper

Overcoming the Barriers of Graphs on GPUs: Delivering Graph Analy;cs 100X Faster and 40X Cheaper Overcoming the Barriers of Graphs on GPUs: Delivering Graph Analy;cs 100X Faster and 40X Cheaper November 18, 2015 Super Compu3ng 2015 The Amount of Graph Data is Exploding! Billion+ Edges! 2 Graph Applications

More information

Convergence of Communication and Machine Learning

Convergence of Communication and Machine Learning Convergence of Communication and Machine Learning Fraunhofer Heinrich Hertz Institute Globally active player in digital infrastructure research Annual budget of 50 M / 450 Researchers Research & Development

More information

arxiv: v1 [cs.cv] 5 Jun 2015

arxiv: v1 [cs.cv] 5 Jun 2015 Beyond Temporal Pooling: Recurrence and Temporal Convolutions for Gesture Recognition in Video arxiv:1506.01911v1 [cs.cv] 5 Jun 2015 Lionel Pigou, Aäron van den Oord, Sander Dieleman, Mieke Van Herreweghe,

More information

Multimodal Medical Image Retrieval based on Latent Topic Modeling

Multimodal Medical Image Retrieval based on Latent Topic Modeling Multimodal Medical Image Retrieval based on Latent Topic Modeling Mandikal Vikram 15it217.vikram@nitk.edu.in Suhas BS 15it110.suhas@nitk.edu.in Aditya Anantharaman 15it201.aditya.a@nitk.edu.in Sowmya Kamath

More information

COMP 551 Applied Machine Learning Lecture 16: Deep Learning

COMP 551 Applied Machine Learning Lecture 16: Deep Learning COMP 551 Applied Machine Learning Lecture 16: Deep Learning Instructor: Ryan Lowe (ryan.lowe@cs.mcgill.ca) Slides mostly by: Class web page: www.cs.mcgill.ca/~hvanho2/comp551 Unless otherwise noted, all

More information

Inception and Residual Networks. Hantao Zhang. Deep Learning with Python.

Inception and Residual Networks. Hantao Zhang. Deep Learning with Python. Inception and Residual Networks Hantao Zhang Deep Learning with Python https://en.wikipedia.org/wiki/residual_neural_network Deep Neural Network Progress from Large Scale Visual Recognition Challenge (ILSVRC)

More information

Why is computer vision difficult?

Why is computer vision difficult? Why is computer vision difficult? Viewpoint variation Illumination Scale Why is computer vision difficult? Intra-class variation Motion (Source: S. Lazebnik) Background clutter Occlusion Challenges: local

More information

Deep Learning Accelerators

Deep Learning Accelerators Deep Learning Accelerators Abhishek Srivastava (as29) Samarth Kulshreshtha (samarth5) University of Illinois, Urbana-Champaign Submitted as a requirement for CS 433 graduate student project Outline Introduction

More information

arxiv: v1 [cs.lg] 16 Jan 2013

arxiv: v1 [cs.lg] 16 Jan 2013 Stochastic Pooling for Regularization of Deep Convolutional Neural Networks arxiv:131.3557v1 [cs.lg] 16 Jan 213 Matthew D. Zeiler Department of Computer Science Courant Institute, New York University zeiler@cs.nyu.edu

More information

Image Classification using Fast Learning Convolutional Neural Networks

Image Classification using Fast Learning Convolutional Neural Networks , pp.50-55 http://dx.doi.org/10.14257/astl.2015.113.11 Image Classification using Fast Learning Convolutional Neural Networks Keonhee Lee 1 and Dong-Chul Park 2 1 Software Device Research Center Korea

More information

CAP 6412 Advanced Computer Vision

CAP 6412 Advanced Computer Vision CAP 6412 Advanced Computer Vision http://www.cs.ucf.edu/~bgong/cap6412.html Boqing Gong April 21st, 2016 Today Administrivia Free parameters in an approach, model, or algorithm? Egocentric videos by Aisha

More information

Real-time convolutional networks for sonar image classification in low-power embedded systems

Real-time convolutional networks for sonar image classification in low-power embedded systems Real-time convolutional networks for sonar image classification in low-power embedded systems Matias Valdenegro-Toro Ocean Systems Laboratory - School of Engineering & Physical Sciences Heriot-Watt University,

More information

Deep Face Recognition. Nathan Sun

Deep Face Recognition. Nathan Sun Deep Face Recognition Nathan Sun Why Facial Recognition? Picture ID or video tracking Higher Security for Facial Recognition Software Immensely useful to police in tracking suspects Your face will be an

More information

Machine Learning. Deep Learning. Eric Xing (and Pengtao Xie) , Fall Lecture 8, October 6, Eric CMU,

Machine Learning. Deep Learning. Eric Xing (and Pengtao Xie) , Fall Lecture 8, October 6, Eric CMU, Machine Learning 10-701, Fall 2015 Deep Learning Eric Xing (and Pengtao Xie) Lecture 8, October 6, 2015 Eric Xing @ CMU, 2015 1 A perennial challenge in computer vision: feature engineering SIFT Spin image

More information