FPGA Implementation of a HOG-based Pedestrian Recognition System
|
|
|
- Brian Powell
- 6 years ago
- Views:
Transcription
1 MPC Workshop Karlsruhe 10/7/2009 FPGA Implementation of a HOG-based Pedestrian Recognition System Sebastian Bauer sebastian.bauer@fh-aschaffenburg.de Laboratory for Pattern Recognition and Computational Intelligence Prof. Dr. Ulrich Brunsmann
2 Content Motivation Method Implementation Evaluation Summary and Outlook 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 2
3 Problem Scope 1,200,000 road traffic deaths per year [WHO/Worldbank 2004] Majority among vulnerable road users (VRU) No advanced driver assistance system (ADAS) available for daylight pedestrian recognition 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 3
4 Problem Scope 1,200,000 road traffic deaths per year [WHO/Worldbank 2004] Majority among vulnerable road users (VRU) No advanced driver assistance system (ADAS) available for daylight pedestrian recognition State-of-the-Art HOG detection system yields best performance [Dollár 2009],[Enzweiler 2009] Challenge: CPU computation time: > 10s (VGA) 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 4
![Problem Scope 1,200,000 road traffic deaths per year [WHO/Worldbank 2004] Majority among vulnerable road users (VRU) No advanced driver assistance system (ADAS) available for daylight pedestrian](/docs-images/79/79443069/images/5-0.jpg "recognition State-of-the-Art HOG detection system yields best performance [Dollár 2009],[Enzweiler 2009] Challenge: CPU computation time: > 10s (VGA) Goal: Real-time HOG system with FPGA 10/7/2009")
5 Problem Scope 1,200,000 road traffic deaths per year [WHO/Worldbank 2004] Majority among vulnerable road users (VRU) No advanced driver assistance system (ADAS) available for daylight pedestrian recognition State-of-the-Art HOG detection system yields best performance [Dollár 2009],[Enzweiler 2009] Challenge: CPU computation time: > 10s (VGA) Goal: Real-time HOG system with FPGA 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 5
6 HOG Detection System Sliding Detection Window For each window: 1. Generate Feature Set / Descriptor (HOG) 2. Classify pedestrian / non-pedestrian 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 6
7 HOG Detection System Sliding Detection Window For each window: 1. Generate Feature Set / Descriptor (HOG) 2. Classify pedestrian / non-pedestrian 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 7
8 HOG Descriptor Histograms of Oriented Gradients [Dalal 2005] Detection Window 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 8
![2005] Gradient Computation](/docs-images/79/79443069/images/9-1.jpg "Point derivatives Magnitude")
9 HOG Descriptor Histograms of Oriented Gradients [Dalal 2005] Gradient Computation Point derivatives Magnitude Orientation 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 9
10 HOG Descriptor Orientation Histogram Detection Window subdivided into Cells 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 10
11 HOG Descriptor Orientation Histogram Detection Window subdivided into Cells For each Cell: Histogram Generation 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 11
12 HOG Descriptor Orientation Histogram Detection Window subdivided into Cells For each Cell: Histogram Generation 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 12
13 HOG Descriptor Orientation Histogram Detection Window subdivided into Cells For each Cell: Histogram Generation 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 13
14 HOG Descriptor Orientation Histogram Detection Window subdivided into Cells For each Cell: Histogram Generation Example: 4x4 px Cell, 3 orientation bins G ϕ Bin 1 Bin 2 Bin /7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 14 Accumulated Magnitude
15 HOG Descriptor Orientation Histogram Detection Window subdivided into Cells For each Cell: Histogram Generation Example: 4x4 px Cell, 3 orientation bins G ϕ Bin 1 Bin 2 Bin /7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 15 Accumulated Magnitude 569
16 HOG Descriptor Orientation Histogram Detection Window subdivided into Cells For each Cell: Histogram Generation Example: 4x4 px Cell, 3 orientation bins G ϕ Bin 1 Bin 2 Bin /7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 16 Accumulated Magnitude
17 HOG Descriptor Orientation Histogram Detection Window subdivided into Cells For each Cell: Histogram Generation Example: 4x4 px Cell, 3 orientation bins G ϕ Bin 1 Bin 2 Bin /7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 17 Accumulated Magnitude
18 HOG Descriptor Block Normalization Block: 2x2 Cells 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 18
19 HOG Descriptor Block Normalization Block: 2x2 Cells For each Block: Concatenation of Cell histograms /7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 19
20 HOG Descriptor Block Normalization Block: 2x2 Cells For each Block: Concatenation of Cell histograms Normalization using L2-Norm /7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 20
21 HOG Descriptor Block Normalization Block: 2x2 Cells For each Block: Concatenation of Cell histograms Normalization using L2-Norm /7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 21
22 HOG Descriptor Block Normalization Block: 2x2 Cells For each Block: Concatenation of Cell histograms Normalization using L2-Norm 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 22
23 HOG Descriptor Block Normalization Block: 2x2 Cells For each Block: Concatenation of Cell histograms Normalization using L2-Norm 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 23
24 HOG Descriptor Block Normalization Block: 2x2 Cells For each Block: Concatenation of Cell histograms Normalization using L2-Norm 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 24
25 HOG Descriptor HOG Descriptor 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 25
26 HOG Descriptor HOG Descriptor Concatenation of normalized Blocks Feature Vector with 3780 dimensions 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 26
27 HOG Descriptor HOG Descriptor Concatenation of normalized Blocks Feature Vector with 3780 dimensions 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 27
28 Classification H O G Classifier Pedestrian Non-Pedestrian Machine Learning Learning from Examples (I/O pairs) Established Pedestrian Datasets [INRIA] 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 28
29 Classification H O G Classifier Pedestrian Non-Pedestrian Machine Learning Learning from Examples (I/O pairs) Established Pedestrian Datasets [INRIA] Support Vector Machines (SVM) Linear Classifier Decision function: y = f(x) = sgn(w x) w learned by training 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 29
30 Classification H O G Classifier Pedestrian Non-Pedestrian Machine Learning Learning from Examples (I/O pairs) Established Pedestrian Datasets [INRIA] Support Vector Machines (SVM) Linear Classifier Decision function: y = f(x) = sgn(w x) w learned by training 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 30
31 Classification H O G Classifier Pedestrian Non-Pedestrian Machine Learning Learning from Examples (I/O pairs) Established Pedestrian Datasets [INRIA] Support Vector Machines (SVM) Linear Classifier Decision function: y = f(x) = sgn(w x) w learned by training 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 31
32 FPGA-CPU-GPU Framework FPGA Xilinx Spartan 3 XC3S 4000 CPU Intel Core i7, 2.66 GHz GPU NVIDIA GeForce GTX /7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 32
![FPGA Rapid Prototyping Platform microenable IV-FULLx4 [SiliconSoftware] Commercially available frame grabber board FPGA for](/docs-images/79/79443069/images/33-0.jpg "individual image pre-processing Camera Link 128 MB 128 MB 128 MB 128 MB Host PCIe 10/7/2009 Sebastian Bauer, MPC Workshop")
33 FPGA Rapid Prototyping Platform microenable IV-FULLx4 [SiliconSoftware] Commercially available frame grabber board FPGA for individual image pre-processing Camera Link 128 MB 128 MB 128 MB 128 MB Host PCIe 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 33
34 FPGA-HOG VisualApplets Graphic-oriented hardware development software, based on Xilinx ISE tools Design arranged by image processing operators and transport links HOG flowchart Reference Magnitude Camera Interface Preprocessing Gradient Cell Histogram DMA Interface Orientation Orientation Binning 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 34
35 FPGA-HOG Magnitude square root extracted, as Euclidean metric yields best performance Orientation Binning Example: Bin 1 (0-20 ) 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 35
36 FPGA-HOG Histogram Generation φ Example: 4x4 px Cell 3 orientation bins Accumulated Magnitude Bin Bin Bin G /7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 36
37 FPGA-HOG Histogram Generation Orientation binning O1: 0-60 O2: O3: φ Example: 4x4 px Cell 3 orientation bins O 1 O 2 O Accumulated Magnitude G 569 Bin Bin Bin /7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 37
38 FPGA-HOG Histogram Generation Orientation binning O1: 0-60 O2: O3: φ Example: 4x4 px Cell 3 orientation bins O 1 O 2 O Accumulated Magnitude G 569 Bin Bin Bin Magnitude weighting M 1 M 2 M /7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 38
39 FPGA-HOG Histogram Generation Orientation binning O1: 0-60 O2: O3: φ Example: 4x4 px Cell 3 orientation bins O 1 O 2 O Accumulated Magnitude G 569 Bin Bin Bin Magnitude weighting M 1 M 2 M K S Histogram bin accumulation B 1 B 2 B /7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 39
40 FPGA-HOG Histogram Generation Orientation binning O1: 0-60 O2: O3: Magnitude weighting φ O 1 O 2 O M 1 M 2 M Example: 4x4 px Cell 3 orientation bins Accumulated Magnitude G K S 569 Bin Bin Bin Histogram bin accumulation B 1 B 2 B /7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 40
41 Classification Performance Evaluation on [INRIA] benchmark dataset Detection Error Tradeoff curves (DET) 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 41
42 Classification Performance Evaluation on [INRIA] benchmark dataset Detection Error Tradeoff curves (DET) Comparison to other detectors close to reference (Ker. R-HOG) without retraining 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 42
43 Time/Throughput Latency HOG step Latency [µs] 800x600 px Image Buffer 26.2 Downsampling 26.0 Gradient 52.0 Magnitude/Orientation 0.1 Histogram TOTAL µs 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 43
44 Time/Throughput Latency HOG step Latency [µs] 800x600 px Image Buffer 26.2 Downsampling 26.0 Gradient 52.0 Magnitude/Orientation 0.1 Histogram TOTAL µs GPU-HOG [Wojek 2008] 13 ms (320x240 px) 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 44
45 Time/Throughput Latency HOG step Latency [µs] 800x600 px Image Buffer 26.2 Downsampling 26.0 Gradient 52.0 Magnitude/Orientation 0.1 Histogram TOTAL µs GPU-HOG [Wojek 2008] 13 ms (320x240 px) Throughput: full 30 fps (1600x1200 px) delivered by camera 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 45
46 Time/Throughput Latency HOG step Latency [µs] 800x600 px Image Buffer 26.2 Downsampling 26.0 Gradient 52.0 Magnitude/Orientation 0.1 Histogram TOTAL µs GPU-HOG [Wojek 2008] 13 ms (320x240 px) Throughput: full 30 fps (1600x1200 px) delivered by camera Resources Resource Type Total Number Usage Level 4 input LUTs 42,435 76% Internal RAM 60 43% Embedded Multipliers 18 18% Xilinx Spartan 3 XC3S 4000 Processing resolution: 800x600 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 46
47 What about our Goals? HOG descriptor in real-time on low cost FPGA Good classification results without retraining Real-time pedestrian recognition system for crossroad assistance (20 fps) 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 47
48 What about our Goals? HOG descriptor in real-time on low cost FPGA Good classification results without retraining Real-time pedestrian recognition system for crossroad assistance (20 fps) Outlook Evaluation with retrained classifier Outsource further parts to FPGA (normalization, SVM) Application for recognition of other traffic participants 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 48
49 Thank you for your attention. If you have any questions, feel free to ask. 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 49
50 References [Dalal::2005] [Cao::2008] [Zhu::2006] Histograms of Oriented Gradients for Human Detection. Dalal, Triggs. CVPR Real-Time Vision-based Stop Sign Detection System on FPGA. Cao, Deng. DICTA 2008 Fast Human Detection Using a Cascade of Histograms of Oriented Gradients. Zhu, Avidan, Yeh, Cheng. CVPR MERL. 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 50
51 Additional Slides VA: Gradient Convolution 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 51
52 Additional Slides VA: Latency Measurement 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 52
53 Additional Slides VA: Gradient Magnitude 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 53
54 Additional Slides How to bypass arctan() 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 54
55 Additional Slides HOG Modifications Grayscale images: 1.5% lower detection rate at 1E-4 FPPW (compared to color images) Gradient magnitude: decimal places are lost when the square root is extracted No Anti-Aliasing: Interpolatation of votes between neighbouring histogram bin centres in both orientation and position No Gaussian down-weighting of pixels near the edges of the block: 1% lower detection rate at 1E-4 FPPW 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 55
56 Additional Slides Solving Nonlinear Problems with SVMs Kernel Trick Decision Function: 10/7/2009 Sebastian Bauer, MPC Workshop Karlsruhe 56
CS4495/6495 Introduction to Computer Vision. 8C-L3 Support Vector Machines
CS4495/6495 Introduction to Computer Vision 8C-L3 Support Vector Machines Discriminative classifiers Discriminative classifiers find a division (surface) in feature space that separates the classes Several
Discriminative part-based models. Many slides based on P. Felzenszwalb
More sliding window detection: ti Discriminative part-based models Many slides based on P. Felzenszwalb Challenge: Generic object detection Pedestrian detection Features: Histograms of oriented gradients
A Discriminatively Trained, Multiscale, Deformable Part Model
A Discriminatively Trained, Multiscale, Deformable Part Model P. Felzenszwalb, D. McAllester, and D. Ramanan Edward Hsiao 16-721 Learning Based Methods in Vision February 16, 2009 Images taken from P.
Maarten Bieshaar, Günther Reitberger, Stefan Zernetsch, Prof. Dr. Bernhard Sick, Dr. Erich Fuchs, Prof. Dr.-Ing. Konrad Doll
Maarten Bieshaar, Günther Reitberger, Stefan Zernetsch, Prof. Dr. Bernhard Sick, Dr. Erich Fuchs, Prof. Dr.-Ing. Konrad Doll 08.02.2017 By 2030 road traffic deaths will be the fifth leading cause of death
DISCRIMINATIVE DECORELATION FOR CLUSTERING AND CLASSIFICATION
DISCRIMINATIVE DECORELATION FOR CLUSTERING AND CLASSIFICATION ECCV 12 Bharath Hariharan, Jitandra Malik, and Deva Ramanan MOTIVATION State-of-the-art Object Detection HOG Linear SVM Dalal&Triggs Histograms
Shape of Gaussians as Feature Descriptors
Shape of Gaussians as Feature Descriptors Liyu Gong, Tianjiang Wang and Fang Liu Intelligent and Distributed Computing Lab, School of Computer Science and Technology Huazhong University of Science and
Real Time Face Detection and Recognition using Haar - Based Cascade Classifier and Principal Component Analysis
Real Time Face Detection and Recognition using Haar - Based Cascade Classifier and Principal Component Analysis Sarala A. Dabhade PG student M. Tech (Computer Egg) BVDU s COE Pune Prof. Mrunal S. Bewoor
One-class Label Propagation Using Local Cone Based Similarity
One-class Label Propagation Using Local Based Similarity Takumi Kobayashi and Nobuyuki Otsu Abstract In this paper, we propose a novel method of label propagation for one-class learning. For binary (positive/negative)
Two-stage Pedestrian Detection Based on Multiple Features and Machine Learning
38 3 Vol. 38, No. 3 2012 3 ACTA AUTOMATICA SINICA March, 2012 1 1 1, (Adaboost) (Support vector machine, SVM). (Four direction features, FDF) GAB (Gentle Adaboost) (Entropy-histograms of oriented gradients,
Max-Margin Additive Classifiers for Detection
Max-Margin Additive Classifiers for Detection Subhransu Maji and Alexander C. Berg Sam Hare VGG Reading Group October 30, 2009 Introduction CVPR08: SVMs with additive kernels can be evaluated efficiently.
Compressed Fisher vectors for LSVR
XRCE@ILSVRC2011 Compressed Fisher vectors for LSVR Florent Perronnin and Jorge Sánchez* Xerox Research Centre Europe (XRCE) *Now with CIII, Cordoba University, Argentina Our system in a nutshell High-dimensional
CS 231A Section 1: Linear Algebra & Probability Review
CS 231A Section 1: Linear Algebra & Probability Review 1 Topics Support Vector Machines Boosting Viola-Jones face detector Linear Algebra Review Notation Operations & Properties Matrix Calculus Probability
CS 231A Section 1: Linear Algebra & Probability Review. Kevin Tang
CS 231A Section 1: Linear Algebra & Probability Review Kevin Tang Kevin Tang Section 1-1 9/30/2011 Topics Support Vector Machines Boosting Viola Jones face detector Linear Algebra Review Notation Operations
Face detection and recognition. Detection Recognition Sally
Face detection and recognition Detection Recognition Sally Face detection & recognition Viola & Jones detector Available in open CV Face recognition Eigenfaces for face recognition Metric learning identification
Workshop on Web- scale Vision and Social Media ECCV 2012, Firenze, Italy October, 2012 Linearized Smooth Additive Classifiers
Worshop on Web- scale Vision and Social Media ECCV 2012, Firenze, Italy October, 2012 Linearized Smooth Additive Classifiers Subhransu Maji Research Assistant Professor Toyota Technological Institute at
OHW2013 workshop. An open source PCIe device virtualization framework
OHW2013 workshop An open source PCIe device virtualization framework Plan Context and objectives Design and implementation Future directions Questions Context - ESRF and the ISDD electronic laboratory
Deep Learning (CNNs)
10-601 Introduction to Machine Learning Machine Learning Department School of Computer Science Carnegie Mellon University Deep Learning (CNNs) Deep Learning Readings: Murphy 28 Bishop - - HTF - - Mitchell
Fast Human Detection from Videos Using Covariance Features
THIS PAPER APPEARED IN THE ECCV VISUAL SURVEILLANCE WORKSHOP (ECCV-VS), MARSEILLE, OCTOBER 2008 Fast Human Detection from Videos Using Covariance Features Jian Yao Jean-Marc Odobez IDIAP Research Institute
Observing Dark Worlds (Final Report)
Observing Dark Worlds (Final Report) Bingrui Joel Li (0009) Abstract Dark matter is hypothesized to account for a large proportion of the universe s total mass. It does not emit or absorb light, making
CS 1674: Intro to Computer Vision. Final Review. Prof. Adriana Kovashka University of Pittsburgh December 7, 2016
CS 1674: Intro to Computer Vision Final Review Prof. Adriana Kovashka University of Pittsburgh December 7, 2016 Final info Format: multiple-choice, true/false, fill in the blank, short answers, apply an
Tennis player segmentation for semantic behavior analysis
Proposta di Tennis player segmentation for semantic behavior analysis Architettura Software per Robot Mobili Vito Renò, Nicola Mosca, Massimiliano Nitti, Tiziana D Orazio, Donato Campagnoli, Andrea Prati,
ML (cont.): SUPPORT VECTOR MACHINES
ML (cont.): SUPPORT VECTOR MACHINES CS540 Bryan R Gibson University of Wisconsin-Madison Slides adapted from those used by Prof. Jerry Zhu, CS540-1 1 / 40 Support Vector Machines (SVMs) The No-Math Version
Pictorial Structures Revisited: People Detection and Articulated Pose Estimation. Department of Computer Science TU Darmstadt
Pictorial Structures Revisited: People Detection and Articulated Pose Estimation Mykhaylo Andriluka Stefan Roth Bernt Schiele Department of Computer Science TU Darmstadt Generic model for human detection
Towards Fully-automated Driving
Towards Fully-automated Driving Challenges and Potential Solutions Dr. Gijs Dubbelman Mobile Perception Systems EE-SPS/VCA Mobile Perception Systems 6 PhDs, postdoc, project manager, software engineer,
Machine Learning for Gravitational Wave signals classification in LIGO and Virgo
Machine Learning for Gravitational Wave signals classification in LIGO and Virgo Elena Cuoco European Gravitational Observatory www.elenacuoco.com @elenacuoco 2 About me About me Working as Data Analyst
Kernel Learning for Multi-modal Recognition Tasks
Kernel Learning for Multi-modal Recognition Tasks J. Saketha Nath CSE, IIT-B IBM Workshop J. Saketha Nath (IIT-B) IBM Workshop 23-Sep-09 1 / 15 Multi-modal Learning Tasks Multiple views or descriptions
Objective Metrics and Gradient Descent Algorithms for Adversarial Examples in Machine Learning
Objective Metrics and Gradient Descent Algorithms for Adversarial Examples in Machine Learning Uyeong Jang Xi Wu Somesh Jha University of Wisconsin Madison, Wisconsin wjang@cs.wisc.edu Google wu.andrew.xi@gmail.com
Optimizing Loop Operation and Dataflow in FPGA Acceleration of Deep Convolutional Neural Networks
2017 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays Optimizing Loop Operation and Dataflow in FPGA Acceleration of Deep Convolutional Neural Networks Yufei Ma, Yu Cao, Sarma Vrudhula,
Learning theory. Ensemble methods. Boosting. Boosting: history
Learning theory Probability distribution P over X {0, 1}; let (X, Y ) P. We get S := {(x i, y i )} n i=1, an iid sample from P. Ensemble methods Goal: Fix ɛ, δ (0, 1). With probability at least 1 δ (over
OBJECT DETECTION AND RECOGNITION IN DIGITAL IMAGES
OBJECT DETECTION AND RECOGNITION IN DIGITAL IMAGES THEORY AND PRACTICE Bogustaw Cyganek AGH University of Science and Technology, Poland WILEY A John Wiley &. Sons, Ltd., Publication Contents Preface Acknowledgements
Robust License Plate Detection Using Covariance Descriptor in a Neural Network Framework
MITSUBISHI ELECTRIC RESEARCH LABORATORIES http://www.merl.com Robust License Plate Detection Using Covariance Descriptor in a Neural Network Framework Fatih Porikli, Tekin Kocak TR2006-100 January 2007
CS 3710: Visual Recognition Describing Images with Features. Adriana Kovashka Department of Computer Science January 8, 2015
CS 3710: Visual Recognition Describing Images with Features Adriana Kovashka Department of Computer Science January 8, 2015 Plan for Today Presentation assignments + schedule changes Image filtering Feature
Support Vector Machines
Support Vector Machines Tobias Pohlen Selected Topics in Human Language Technology and Pattern Recognition February 10, 2014 Human Language Technology and Pattern Recognition Lehrstuhl für Informatik 6
Fisher Vector image representation
Fisher Vector image representation Machine Learning and Category Representation 2014-2015 Jakob Verbeek, January 9, 2015 Course website: http://lear.inrialpes.fr/~verbeek/mlcr.14.15 A brief recap on kernel
Loss Functions and Optimization. Lecture 3-1
Lecture 3: Loss Functions and Optimization Lecture 3-1 Administrative Assignment 1 is released: http://cs231n.github.io/assignments2017/assignment1/ Due Thursday April 20, 11:59pm on Canvas (Extending
Fantope Regularization in Metric Learning
Fantope Regularization in Metric Learning CVPR 2014 Marc T. Law (LIP6, UPMC), Nicolas Thome (LIP6 - UPMC Sorbonne Universités), Matthieu Cord (LIP6 - UPMC Sorbonne Universités), Paris, France Introduction
Neural Networks. Prof. Dr. Rudolf Kruse. Computational Intelligence Group Faculty for Computer Science
Neural Networks Prof. Dr. Rudolf Kruse Computational Intelligence Group Faculty for Computer Science kruse@iws.cs.uni-magdeburg.de Rudolf Kruse Neural Networks 1 Supervised Learning / Support Vector Machines
Implementation of Nonlinear Template Runner Emulated Digital CNN-UM on FPGA
Implementation of Nonlinear Template Runner Emulated Digital CNN-UM on FPGA Z. Kincses * Z. Nagy P. Szolgay * Department of Image Processing and Neurocomputing, University of Pannonia, Hungary * e-mail:
MATHEMATICAL entities that do not form Euclidean
Kernel Methods on Riemannian Manifolds with Gaussian RBF Kernels Sadeep Jayasumana, Student Member, IEEE, Richard Hartley, Fellow, IEEE, Mathieu Salzmann, Member, IEEE, Hongdong Li, Member, IEEE, and Mehrtash
Properties of detectors Edge detectors Harris DoG Properties of descriptors SIFT HOG Shape context
Lecture 10 Detectors and descriptors Properties of detectors Edge detectors Harris DoG Properties of descriptors SIFT HOG Shape context Silvio Savarese Lecture 10-16-Feb-15 From the 3D to 2D & vice versa
Support Vector Machines
Support Vector Machines INFO-4604, Applied Machine Learning University of Colorado Boulder September 28, 2017 Prof. Michael Paul Today Two important concepts: Margins Kernels Large Margin Classification
Edges and Scale. Image Features. Detecting edges. Origin of Edges. Solution: smooth first. Effects of noise
Edges and Scale Image Features From Sandlot Science Slides revised from S. Seitz, R. Szeliski, S. Lazebnik, etc. Origin of Edges surface normal discontinuity depth discontinuity surface color discontinuity
Online Multi-Object Tracking with Dual Matching Attention Networks
Online Multi-Object Tracking with Dual Matching Attention Networks Ji Zhu 1,2, Hua Yang 1, Nian Liu 3, Minyoung Kim 4, Wenjun Zhang 1, and Ming-Hsuan Yang 5,6 1 Shanghai Jiao Tong University 2 Visbody
Kernel Methods on the Riemannian Manifold of Symmetric Positive Definite Matrices
2013 IEEE Conference on Computer ision and Pattern Recognition Kernel Methods on the Riemannian Manifold of Symmetric Positive Definite Matrices Sadeep Jayasumana 1, 2, Richard Hartley 1, 2, Mathieu Salzmann
Vector Quantization for Machine Vision
1 Vector Quantization for Machine Vision Vincenzo Liguori Ocean Logic Pty Ltd, Australia Email: enzo@ocean-logic.com Abstract This paper shows how to reduce the computational cost for a variety of common
CS5670: Computer Vision
CS5670: Computer Vision Noah Snavely Lecture 5: Feature descriptors and matching Szeliski: 4.1 Reading Announcements Project 1 Artifacts due tomorrow, Friday 2/17, at 11:59pm Project 2 will be released
Classification of Hand-Written Digits Using Scattering Convolutional Network
Mid-year Progress Report Classification of Hand-Written Digits Using Scattering Convolutional Network Dongmian Zou Advisor: Professor Radu Balan Co-Advisor: Dr. Maneesh Singh (SRI) Background Overview
Support Vector Machines. Introduction to Data Mining, 2 nd Edition by Tan, Steinbach, Karpatne, Kumar
Data Mining Support Vector Machines Introduction to Data Mining, 2 nd Edition by Tan, Steinbach, Karpatne, Kumar 02/03/2018 Introduction to Data Mining 1 Support Vector Machines Find a linear hyperplane
Achieving scale covariance
Achieving scale covariance Goal: independently detect corresponding regions in scaled versions of the same image Need scale selection mechanism for finding characteristic region size that is covariant
Support Vector Machines: Maximum Margin Classifiers
Support Vector Machines: Maximum Margin Classifiers Machine Learning and Pattern Recognition: September 16, 2008 Piotr Mirowski Based on slides by Sumit Chopra and Fu-Jie Huang 1 Outline What is behind
The state of the art and beyond
Feature Detectors and Descriptors The state of the art and beyond Local covariant detectors and descriptors have been successful in many applications Registration Stereo vision Motion estimation Matching
Learning Methods for Linear Detectors
Intelligent Systems: Reasoning and Recognition James L. Crowley ENSIMAG 2 / MoSIG M1 Second Semester 2011/2012 Lesson 20 27 April 2012 Contents Learning Methods for Linear Detectors Learning Linear Detectors...2
Dynamic Data Modeling, Recognition, and Synthesis. Rui Zhao Thesis Defense Advisor: Professor Qiang Ji
Dynamic Data Modeling, Recognition, and Synthesis Rui Zhao Thesis Defense Advisor: Professor Qiang Ji Contents Introduction Related Work Dynamic Data Modeling & Analysis Temporal localization Insufficient
Deep Learning. Convolutional Neural Network (CNNs) Ali Ghodsi. October 30, Slides are partially based on Book in preparation, Deep Learning
Convolutional Neural Network (CNNs) University of Waterloo October 30, 2015 Slides are partially based on Book in preparation, by Bengio, Goodfellow, and Aaron Courville, 2015 Convolutional Networks Convolutional
CITS 4402 Computer Vision
CITS 4402 Computer Vision A/Prof Ajmal Mian Adj/A/Prof Mehdi Ravanbakhsh Lecture 06 Object Recognition Objectives To understand the concept of image based object recognition To learn how to match images
Neural Networks. Single-layer neural network. CSE 446: Machine Learning Emily Fox University of Washington March 10, /9/17
3/9/7 Neural Networks Emily Fox University of Washington March 0, 207 Slides adapted from Ali Farhadi (via Carlos Guestrin and Luke Zettlemoyer) Single-layer neural network 3/9/7 Perceptron as a neural
LoG Blob Finding and Scale. Scale Selection. Blobs (and scale selection) Achieving scale covariance. Blob detection in 2D. Blob detection in 2D
Achieving scale covariance Blobs (and scale selection) Goal: independently detect corresponding regions in scaled versions of the same image Need scale selection mechanism for finding characteristic region
Sparse Kernel Machines - SVM
Sparse Kernel Machines - SVM Henrik I. Christensen Robotics & Intelligent Machines @ GT Georgia Institute of Technology, Atlanta, GA 30332-0280 hic@cc.gatech.edu Henrik I. Christensen (RIM@GT) Support
Jeff Howbert Introduction to Machine Learning Winter
Classification / Regression Support Vector Machines Jeff Howbert Introduction to Machine Learning Winter 2012 1 Topics SVM classifiers for linearly separable classes SVM classifiers for non-linearly separable
Introduction to Deep Learning CMPT 733. Steven Bergner
Introduction to Deep Learning CMPT 733 Steven Bergner Overview Renaissance of artificial neural networks Representation learning vs feature engineering Background Linear Algebra, Optimization Regularization
Accelerating linear algebra computations with hybrid GPU-multicore systems.
Accelerating linear algebra computations with hybrid GPU-multicore systems. Marc Baboulin INRIA/Université Paris-Sud joint work with Jack Dongarra (University of Tennessee and Oak Ridge National Laboratory)
Linear Support Vector Machine. Classification. Linear SVM. Huiping Cao. Huiping Cao, Slide 1/26
Huiping Cao, Slide 1/26 Classification Linear SVM Huiping Cao linear hyperplane (decision boundary) that will separate the data Huiping Cao, Slide 2/26 Support Vector Machines rt Vector Find a linear Machines
Linear Classifiers as Pattern Detectors
Intelligent Systems: Reasoning and Recognition James L. Crowley ENSIMAG 2 / MoSIG M1 Second Semester 2014/2015 Lesson 16 8 April 2015 Contents Linear Classifiers as Pattern Detectors Notation...2 Linear
CSC 411 Lecture 17: Support Vector Machine
CSC 411 Lecture 17: Support Vector Machine Ethan Fetaya, James Lucas and Emad Andrews University of Toronto CSC411 Lec17 1 / 1 Today Max-margin classification SVM Hard SVM Duality Soft SVM CSC411 Lec17
Machine Learning Basics
Security and Fairness of Deep Learning Machine Learning Basics Anupam Datta CMU Spring 2019 Image Classification Image Classification Image classification pipeline Input: A training set of N images, each
Machine Learning Support Vector Machines. Prof. Matteo Matteucci
Machine Learning Support Vector Machines Prof. Matteo Matteucci Discriminative vs. Generative Approaches 2 o Generative approach: we derived the classifier from some generative hypothesis about the way
Machine Learning with Tensor Networks
Machine Learning with Tensor Networks E.M. Stoudenmire and David J. Schwab Advances in Neural Information Processing 29 arxiv:1605.05775 Beijing Jun 2017 Machine learning has physics in its DNA # " # #
Holdout and Cross-Validation Methods Overfitting Avoidance
Holdout and Cross-Validation Methods Overfitting Avoidance Decision Trees Reduce error pruning Cost-complexity pruning Neural Networks Early stopping Adjusting Regularizers via Cross-Validation Nearest
Boosting: Algorithms and Applications
Boosting: Algorithms and Applications Lecture 11, ENGN 4522/6520, Statistical Pattern Recognition and Its Applications in Computer Vision ANU 2 nd Semester, 2008 Chunhua Shen, NICTA/RSISE Boosting Definition
Classification with Perceptrons. Reading:
Classification with Perceptrons Reading: Chapters 1-3 of Michael Nielsen's online book on neural networks covers the basics of perceptrons and multilayer neural networks We will cover material in Chapters
38 1 Vol. 38, No ACTA AUTOMATICA SINICA January, Bag-of-phrases.. Image Representation Using Bag-of-phrases
38 1 Vol. 38, No. 1 2012 1 ACTA AUTOMATICA SINICA January, 2012 Bag-of-phrases 1, 2 1 1 1, Bag-of-words,,, Bag-of-words, Bag-of-phrases, Bag-of-words DOI,, Bag-of-words, Bag-of-phrases, SIFT 10.3724/SP.J.1004.2012.00046
The z-vertex Trigger for Belle II
The z-vertex Trigger for Belle II Sebastian Skambraks Technische Universität München IMPRS Young Scientist Workshop at Ringberg Castle July 10, 2015 1 / 13 Outline Introduction Motivation Signal Flow Multi
Feature extraction: Corners and blobs
Feature extraction: Corners and blobs Review: Linear filtering and edge detection Name two different kinds of image noise Name a non-linear smoothing filter What advantages does median filtering have over
Deep Learning. Convolutional Neural Networks Applications
Deep Learning Using a Convolutional Neural Network Dr. Ing. Morris Riedel Adjunct Associated Professor School of Engineering and Natural Sciences, University of Iceland Research Group Leader, Juelich Supercomputing
CS798: Selected topics in Machine Learning
CS798: Selected topics in Machine Learning Support Vector Machine Jakramate Bootkrajang Department of Computer Science Chiang Mai University Jakramate Bootkrajang CS798: Selected topics in Machine Learning
Urban land use information retrieval based on scene classification of Google Street View images
Urban land use information retrieval based on scene classification of Google Street View images Xiaojiang Li 1, Chuanrong Zhang 1 1 Department of Geography, University of Connecticut, Storrs Email: {xiaojiang.li;chuanrong.zhang}@uconn.edu
Real-time image-based parking occupancy detection using deep learning. Debaditya Acharya, Weilin Yan & Kourosh Khoshelham The University of Melbourne
Real-time image-based parking occupancy detection using deep learning Debaditya Acharya, Weilin Yan & Kourosh Khoshelham The University of Melbourne Slide 1/20 Prologue People spend on Does average that
Support Vector Machines
Wien, June, 2010 Paul Hofmarcher, Stefan Theussl, WU Wien Hofmarcher/Theussl SVM 1/21 Linear Separable Separating Hyperplanes Non-Linear Separable Soft-Margin Hyperplanes Hofmarcher/Theussl SVM 2/21 (SVM)
Active Detection via Adaptive Submodularity
Active Detection via Adaptive Submodularity Yuxin Chen, Hiroaki Shioi, Cesar Antonio Fuentes Montesinos! Lian Pin Koh, Serge Wich and Andreas Krause! ICML Beijing June 23, 2014! Motivating Example: Biodiversity
Wavelet-based Salient Points with Scale Information for Classification
Wavelet-based Salient Points with Scale Information for Classification Alexandra Teynor and Hans Burkhardt Department of Computer Science, Albert-Ludwigs-Universität Freiburg, Germany {teynor, Hans.Burkhardt}@informatik.uni-freiburg.de
Deep Reinforcement Learning SISL. Jeremy Morton (jmorton2) November 7, Stanford Intelligent Systems Laboratory
Deep Reinforcement Learning Jeremy Morton (jmorton2) November 7, 2016 SISL Stanford Intelligent Systems Laboratory Overview 2 1 Motivation 2 Neural Networks 3 Deep Reinforcement Learning 4 Deep Learning
EECS150 - Digital Design Lecture 15 SIFT2 + FSM. Recap and Outline
EECS150 - Digital Design Lecture 15 SIFT2 + FSM Oct. 15, 2013 Prof. Ronald Fearing Electrical Engineering and Computer Sciences University of California, Berkeley (slides courtesy of Prof. John Wawrzynek)
Multimodal context analysis and prediction
Multimodal context analysis and prediction Valeria Tomaselli (valeria.tomaselli@st.com) Sebastiano Battiato Giovanni Maria Farinella Tiziana Rotondo (PhD student) Outline 2 Context analysis vs prediction
Large-scale classification of traffic signs under real-world conditions
Large-scale classification of traffic signs under real-world conditions Lykele Hazelhoff a,b, Ivo Creusen a,b, Dennis van de Wouw a,b and Peter H.N. de With a,b a CycloMedia Technology B.V., Achterweg
Support Vector Machine (SVM) and Kernel Methods
Support Vector Machine (SVM) and Kernel Methods CE-717: Machine Learning Sharif University of Technology Fall 2014 Soleymani Outline Margin concept Hard-Margin SVM Soft-Margin SVM Dual Problems of Hard-Margin
FPGA Implementation of a Predictive Controller
FPGA Implementation of a Predictive Controller SIAM Conference on Optimization 2011, Darmstadt, Germany Minisymposium on embedded optimization Juan L. Jerez, George A. Constantinides and Eric C. Kerrigan
Machine learning comes from Bayesian decision theory in statistics. There we want to minimize the expected value of the loss function.
Bayesian learning: Machine learning comes from Bayesian decision theory in statistics. There we want to minimize the expected value of the loss function. Let y be the true label and y be the predicted
Multi-scale Geometric Summaries for Similarity-based Upstream S
Multi-scale Geometric Summaries for Similarity-based Upstream Sensor Fusion Duke University, ECE / Math 3/6/2019 Overall Goals / Design Choices Leverage multiple, heterogeneous modalities in identification
Kaggle.
Administrivia Mini-project 2 due April 7, in class implement multi-class reductions, naive bayes, kernel perceptron, multi-class logistic regression and two layer neural networks training set: Project
Multivariate statistical methods and data mining in particle physics
Multivariate statistical methods and data mining in particle physics RHUL Physics www.pp.rhul.ac.uk/~cowan Academic Training Lectures CERN 16 19 June, 2008 1 Outline Statement of the problem Some general
Mitosis Detection in Breast Cancer Histology Images with Multi Column Deep Neural Networks
Mitosis Detection in Breast Cancer Histology Images with Multi Column Deep Neural Networks IDSIA, Lugano, Switzerland dan.ciresan@gmail.com Dan C. Cireşan and Alessandro Giusti DNN for Visual Pattern Recognition
Advances in Computer Vision. Prof. Bill Freeman. Image and shape descriptors. Readings: Mikolajczyk and Schmid; Belongie et al.
6.869 Advances in Computer Vision Prof. Bill Freeman March 3, 2005 Image and shape descriptors Affine invariant features Comparison of feature descriptors Shape context Readings: Mikolajczyk and Schmid;
Bayesian Classifiers and Probability Estimation. Vassilis Athitsos CSE 4308/5360: Artificial Intelligence I University of Texas at Arlington
Bayesian Classifiers and Probability Estimation Vassilis Athitsos CSE 4308/5360: Artificial Intelligence I University of Texas at Arlington 1 Data Space Suppose that we have a classification problem The
Explaining Machine Learning Decisions
Explaining Machine Learning Decisions Grégoire Montavon, TU Berlin Joint work with: Wojciech Samek, Klaus-Robert Müller, Sebastian Lapuschkin, Alexander Binder 18/09/2018 Intl. Workshop ML & AI, Telecom
Instance-based Learning CE-717: Machine Learning Sharif University of Technology. M. Soleymani Fall 2016
Instance-based Learning CE-717: Machine Learning Sharif University of Technology M. Soleymani Fall 2016 Outline Non-parametric approach Unsupervised: Non-parametric density estimation Parzen Windows Kn-Nearest
Detecting Humans via Their Pose
Detecting Humans via Their Pose Alessandro Bissacco Computer Science Department University of California, Los Angeles Los Angeles, CA 90095 bissacco@cs.ucla.edu Ming-Hsuan Yang Honda Research Institute
COMPUTER vision and machine intelligence have been
IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 23, NO. 1, JANUARY 2014 287 Human Detection by Quadratic Classification on Subspace of Extended Histogram of Gradients Amit Satpathy, Student Member, IEEE, Xudong
Faster Machine Learning via Low-Precision Communication & Computation. Dan Alistarh (IST Austria & ETH Zurich), Hantian Zhang (ETH Zurich)
Faster Machine Learning via Low-Precision Communication & Computation Dan Alistarh (IST Austria & ETH Zurich), Hantian Zhang (ETH Zurich) 2 How many bits do you need to represent a single number in machine
DETECTING HUMAN ACTIVITIES IN THE ARCTIC OCEAN BY CONSTRUCTING AND ANALYZING SUPER-RESOLUTION IMAGES FROM MODIS DATA INTRODUCTION
DETECTING HUMAN ACTIVITIES IN THE ARCTIC OCEAN BY CONSTRUCTING AND ANALYZING SUPER-RESOLUTION IMAGES FROM MODIS DATA Shizhi Chen and YingLi Tian Department of Electrical Engineering The City College of
A Kernel on Persistence Diagrams for Machine Learning
A Kernel on Persistence Diagrams for Machine Learning Jan Reininghaus 1 Stefan Huber 1 Roland Kwitt 2 Ulrich Bauer 1 1 Institute of Science and Technology Austria 2 FB Computer Science Universität Salzburg,