Scribd Logo
Upload

Welcome to Scribd!

  • 124 views

    Convolutional Neural Networks For Visual Recognition

    Uploaded by

    Keren Evangeline. I
    This document provides an overview and agenda for a course on convolutional neural networks for visual recognition. It introduces convolutional networks and their applications in computer vision tasks like image classification, object detection, and scene parsing. It discusses how convolutional networks have achieved state-of-the-art results in large-scale visual recognition challenges by learning hierarchical representations from data. The document also examines why deep convolutional networks are effective and outlines the course content which will cover network architectures, training, and applications in localization and detection.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd

    Convolutional Neural Networks For Visual Recognition

    Uploaded by

    Keren Evangeline. I
    124 views45 pages
    This document provides an overview and agenda for a course on convolutional neural networks for visual recognition. It introduces convolutional networks and their applications in computer vision tasks like image classification, object detection, and scene parsing. It discusses how convolutional networks have achieved state-of-the-art results in large-scale visual recognition challenges by learning hierarchical representations from data. The document also examines why deep convolutional networks are effective and outlines the course content which will cover network architectures, training, and applications in localization and detection.

    Original Description:

    CNN

    Original Title

    Lecture 1 CNN Introduction

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document provides an overview and agenda for a course on convolutional neural networks for visual recognition. It introduces convolutional networks and their applications in computer vision tasks like image classification, object detection, and scene parsing. It discusses how convolutional networks have achieved state-of-the-art results in large-scale visual recognition challenges by learning hierarchical representations from data. The document also examines why deep convolutional networks are effective and outlines the course content which will cover network architectures, training, and applications in localization and detection.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    124 views45 pages

    Convolutional Neural Networks For Visual Recognition

    Uploaded by

    Keren Evangeline. I
    This document provides an overview and agenda for a course on convolutional neural networks for visual recognition. It introduces convolutional networks and their applications in computer vision tasks like image classification, object detection, and scene parsing. It discusses how convolutional networks have achieved state-of-the-art results in large-scale visual recognition challenges by learning hierarchical representations from data. The document also examines why deep convolutional networks are effective and outlines the course content which will cover network architectures, training, and applications in localization and detection.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    of 45
    At a glance
    Powered by AI
    The presentation discusses the history and success of convolutional neural networks in computer vision tasks. CNNs have achieved human-level performance on tasks like ImageNet classification by learning hierarchical image features automatically from large datasets.

    CNNs achieved breakthrough results on the ImageNet visual recognition challenge in 2012 by significantly outperforming other methods. Their success was due to having many layers that learned increasingly complex features, from edges to shapes to objects.

    CNNs have been applied to tasks beyond classification like detection, scene parsing, indoor semantic labeling, and action detection by adapting the network architecture. They have proven effective for a wide range of computer vision problems.

    Introduction:

    Convolutional Neural Networks


    for Visual Recognition

    boris [email protected]
    1
    Acknowledgments

    This presentation is heavily based on:


    – http://cs.nyu.edu/~fergus/pmwiki/pmwiki.php
    – http://deeplearning.net/reading-list/tutorials/
    – http://deeplearning.net/tutorial/lenet.html
    – http://ufldl.stanford.edu/wiki/index.php/UFLDL_Tutorial

    … and many other

    2
    Agenda

    1. Course overview
    2. Introduction to Deep Learning
    – Classical Computer Vision vs. Deep learning
    3. Introduction to Convolutional Networks
    – Basic CNN Architecture
    – Large Scale Image Classifications
    – How deep should be Conv Nets?
    – Detection and Other Visual Apps

    3
    Course overview

    1. Introduction
    – Intro to Deep Learning
    – Caffe: Getting started
    – CNN: network topology, layers definition
    2. CNN Training
    – Backward propagation
    – Optimization for Deep Learning: SGD : monentum, rate
    adaptation, Adagrad, SGD with Line Search, CGD
    – “Regularization” (Dropout , Maxout)

    4
    Course overview

    3. Localization and Detection


    – Overfeat
    – R-CNN (Regions with CNN)
    4. CPU / GPU performance optimization
    – CUDA
    – Vtune, OpenMP, and Intel MKL (Math Kernel Library)

    5
    Introduction to Deep Learning

    6
    Buzz…

    7
    Deep Learning – from Research to
    Technology

    Deep Learning - breakthrough in


    visual and speech recognition 8
    Classical Computer Vision Pipeline

    9
    Classical Computer Vision Pipeline.

    CV experts
    1. Select / develop features: SURF, HoG, SIFT, RIFT,
    …
    2. Add on top of this Machine Learning for multi-class
    recognition and train classifier
    Feature Detection,
    Extraction: Classification
    SIFT, HoG... Recognition

    Classical CV feature definition is domain-


    specific and time-consuming

    10
    Deep Learning –based Vision Pipeline.

    Deep Learning:
     Build features automatically based on training data
     Combine feature extraction and classification
    DL experts: define NN topology and train NN

    Detection,
    Deep NN... Deep NN...
    Classification
    Recognition

    Deep Learning promise:


    train good feature automatically,
    same method for different domain
    11
    Computer Vision +Deep Learning +
    Machine Learning
    We want to combine Deep Learning + CV + ML
     Combine pre-defined features with learned features;
     Use best ML methods for multi-class recognition
    CV+DL+ML experts needed to build the best-in-class

    CV ML
    Deep AdaBoost
    features
    NN... …
    HoG, SIFT

    Combine best of Computer Vision


    Deep Learning and Machine Learning

    12
    Deep Learning Basics
    Deep Learning – is a set of machine learning
    algorithms based on multi-layer networks
    CAT DOG

    OUTPUTS

    HIDDEN
    NODES

    INPUTS
    13
    Deep Learning Basics
    Deep Learning – is a set of machine learning
    algorithms based on multi-layer networks
    CAT DOG

    Training

    14

    1
    Deep Learning Basics
    Deep Learning – is a set of machine learning
    algorithms based on multi-layer networks
    CAT DOG

    15

    1
    Deep Learning Basics
    Deep Learning – is a set of machine learning
    algorithms based on multi-layer networks
    CAT DOG

    16
    Deep Learning Taxonomy

    Supervised:
    –Convolutional NN ( LeCun)
    –Recurrent Neural nets (Schmidhuber )

    Unsupervised
    –Deep Belief Nets / Stacked RBMs (Hinton)
    –Stacked denoising autoencoders (Bengio)
    –Sparse AutoEncoders ( LeCun, A. Ng, )

    17
    Convolutional Networks

    18
    Convolutional NN

    Convolutional Neural Networks is extension of


    traditional Multi-layer Perceptron, based on 3 ideas:
    1. Local receive fields
    2. Shared weights
    3. Spatial / temporal sub-sampling
    See LeCun paper (1998) on text recognition:
    http://yann.lecun.com/exdb/publis/pdf/lecun-01a.pdf

    19
    What is Convolutional NN ?
    CNN - multi-layer NN architecture
    – Convolutional + Non-Linear Layer
    – Sub-sampling Layer
    – Convolutional +Non-L inear Layer
    – Fully connected layers
     Supervised

    Classi-
    Feature Extraction
    fication

    20
    What is Convolutional NN ?

    2x2

    Convolution + NL Sub-sampling Convolution + NL


    21
    CNN success story: ILSVRC 2012

    Imagenet data base: 14 mln labeled images, 20K categories

    22
    ILSVRC: Classification

    23
    Imagenet Classifications 2012

    24
    ILSVRC 2012: top rankers

    http://www.image-net.org/challenges/LSVRC/2012/results.html

    N Error-5 Algorithm Team Authors


    1 0.153 Deep Conv. Neural Univ. of Krizhevsky et al
    Network Toronto
    2 0.262 Features + Fisher ISI Gunji et al
    Vectors + Linear
    classifier
    3 0.270 Features + FV + SVM OXFORD_VG Simonyan et al
    G
    4 0.271 SIFT + FV + PQ + SVM XRCE/INRIA Perronin et al
    5 0.300 Color desc. + SVM Univ. of van de Sande et
    Amsterdam al

    25
    Imagenet 2013: top rankers

    http://www.image-net.org/challenges/LSVRC/2013/results.php

    N Error-5 Algorithm Team Authors


    1 0.117 Deep Convolutional Clarifi Zeiler
    Neural Network
    2 0.129 Deep Convolutional Nat.Univ Min LIN
    Neural Networks Singapore
    3 0.135 Deep Convolutional NYU Zeiler
    Neural Networks Fergus
    4 0.135 Deep Convolutional Andrew Howard
    Neural Networks
    5 0.137 Deep Convolutional Overfeat Pierre Sermanet
    Neural Networks NYU et al

    26
    Imagenet Classifications 2013

    27
    Conv Net Topology

     5 convolutional layers
     3 fully connected layers + soft-max
     650K neurons , 60 Mln weights

    28
    Why ConvNet should be Deep?

    Rob Fergus, NIPS 2013 29


    Why ConvNet should be Deep?

    30
    Why ConvNet should be Deep?

    31
    Why ConvNet should be Deep?

    32
    Why ConvNet should be Deep?

    33
    Conv Nets:
    beyond Visual Classification

    34
    CNN applications

    CNN is a big Plenty low hanging fruits


    hammer

    You need just a right nail! 35


    Conv NN: Detection

    Sermanet, CVPR 2014


    36
    Conv NN: Scene parsing

    Farabet, PAMI 2013


    37
    CNN: indoor semantic labeling RGBD

    Farabet, 2013
    38
    Conv NN: Action Detection

    Taylor, ECCV 2010


    39
    Conv NN: Image Processing

    Eigen , ICCV 2010


    40
    BACKUP

    BUZZ

    41
    A lot of buzz about Deep Learning

     July 2012 - Started DL lab


     Nov 2012- Big improvement in Speech, OCR:
    – Speech – reduce Error Rate by 25%
    – OCR – reduce Error rate by 30%
     2013 launched 5 DL based products
    – Voice search
    – Photo Wonder
    – Visual search

    42
    A lot of buzz about Deep Learning

    Microsoft On Deep Learning for Speech goto 3:00-5:10


    43
    A lot of buzz about Deep Learning

    Why Google invest in Deep Learning


    44
    A lot of buzz about Deep Learning

    NYU “Deep Learning” Professor LeCun Will Head


    Facebook’s New Artificial Intelligence Lab, Dec 10,
    2013
    45

    You might also like