SlideShare a Scribd company logo
A Fast Content-Based
Image Retrieval Method
Using Deep Visual Features
Hiroki Tanioka (Assist. Prof. Ph.D.)
Tokushima University, Japan
HELLO!
I am Hiroki Tanioka.
You can find me at @taniokah
2
“Quotations are why we
cannot use cosine
similarity on inverted
index-based search
engine, despite cosine
similarity is commonly
used for many
classification problems.”
3
Scoring Method
Includes normalizing every
dot product score with L2
norm of each indexed
document.
THERE ARE TWO REASONS FOR THAT
Higher Dimension
Needs much cost of
calculating dot product
between query vector and
indexed document vector.
4
Calculation
Faster
Dimension
Lower
Small Ideas
My key concept is using normalized index
and truncating feature vectors.
5
1.
FASTER
COSINE SMILARITY
Let’s start making index with
normalized vectors in advance.
COSINE SIMILARITY
TRANSFORM
7
BEFORE:
AFTER:
2.
LOWER
VECTOR DIMENSTION
Let’s reduce features un-preferable
document features for score.
SORT AND REDUCE
FEATURE VECTOR
9
・
・
・
Dataset Statistics
Number of Images: 25,000
Type: Dogs and Cats
Source: Kaggle
Search Engine
Elasticsearch 7.0.1
INFROMATION OF
EXPERIMENTAL ENVELONMENT
Image Features
Model: VGG-16
Number of Features: 1,000
Score: Probability
Machine Specification
MacBook Pro, 2.5 GHz, 16GB
(Retina, 15-inch, Mid 2015)
10
THREE TEST PATERNS
FOR EVALUATION OF
MY SYSTEM
Finding a picture
with original picture
Finding a picture
with low-resolution
Finding a picture
with part of picture
11
AVERAGE RESPONSE TIME
Feature 1 2 3 4 5 10 20 30 40 50 100 200 400 1000
dot-prod 0.23 0.33 0.25 0.36 0.37 0.42 0.74 0.69 0.80 0.86 1.43 2.02 2.86 2.87
L1 0.27 0.27 0.30 0.31 0.35 0.35 0.57 0.80 0.70 0.72 1.35 1.58 2.58 2.85
L2 0.26 0.28 0.28 0.30 0.29 0.35 0.51 0.63 0.80 0.99 1.97 3.80 5.78 6.56
cosine 0.23 0.25 0.27 0.28 0.30 0.35 0.64 0.62 0.61 0.59 1.14 1.43 2.60 3.08
dot+cos 44.0 42.2 39.7 36.1 35.0 36.1 35.4 35.0 37.9 41.2 47.0 46.5 52.0 53.0
12
FAST
ENOUGH
13
Comparison MAP of feature number in 100 top-k results and 1.0 resolution rate.
GOOD
ENOUGH
14
Comparison MAP of resolution rates, in 1,000 feature number and 100 top-k
results.
FAIR
ENOUGH
15
Comparison MAP of feature numbers using quartered partial image query, in
1,000 feature number and 100 top-k results.
FAVORABLE
Pre-Normalizing
Is effective for calculating
cosine similarity faster
under 100 features.
THERE ARE TWO RESULTS
Reducing Features
Is useful, but at least 8
features are required for our
image search experiment.
16
INSTRUCTIONS FOR USE
USE XPACK IN ELASTICSEARCH
https://www.elastic.co/blog/text-
similarity-search-with-vectors-in-
elasticsearch
Elasticsearch 7.3 released
cosine_similarity in xpack.
Still, L1 and L2 are under
construction.
VISIT TO MY GITHUB PAGE
https://github.com/taniokah/icdar-
wml-2019
I’ve just already left my messy
scripts on Github.
I will explain an instruction for my
project.
17
18
THANKS!
Any questions?
You can find me at:
❑ @taniokah
❑ tanioka.hiroki@tokushima-u.ac.jp
This work was supported by JSPS KAKENHI Grant Number JP18H03344.

More Related Content

What's hot

G. Park, J.-Y. Yang, et. al., NeurIPS 2020, MLILAB, KAIST AI
G. Park, J.-Y. Yang, et. al., NeurIPS 2020, MLILAB, KAIST AIG. Park, J.-Y. Yang, et. al., NeurIPS 2020, MLILAB, KAIST AI
G. Park, J.-Y. Yang, et. al., NeurIPS 2020, MLILAB, KAIST AI
MLILAB
 
J. Park, H. Shim, AAAI 2022, MLILAB, KAISTAI
J. Park, H. Shim, AAAI 2022, MLILAB, KAISTAIJ. Park, H. Shim, AAAI 2022, MLILAB, KAISTAI
J. Park, H. Shim, AAAI 2022, MLILAB, KAISTAI
MLILAB
 
IMAGE ENHANCEMENT /ENHANCHING the feature of an image/ Image editor
IMAGE ENHANCEMENT /ENHANCHING the feature of an image/ Image editorIMAGE ENHANCEMENT /ENHANCHING the feature of an image/ Image editor
IMAGE ENHANCEMENT /ENHANCHING the feature of an image/ Image editor
RAHUL DANGWAL
 
Scale Invariant Feature Transform
Scale Invariant Feature TransformScale Invariant Feature Transform
Scale Invariant Feature Transform
kislayabhi
 
Hog
HogHog
The Error Bound on SLLN
The Error Bound on SLLNThe Error Bound on SLLN
The Error Bound on SLLNguest8e0f17
 
The Error Bound on SLLN
The Error Bound on SLLNThe Error Bound on SLLN
The Error Bound on SLLN
guest8e0f17
 
EMBC'13 Poster Presentation on "A Bio-Inspired Cooperative Algorithm for Dist...
EMBC'13 Poster Presentation on "A Bio-Inspired Cooperative Algorithm for Dist...EMBC'13 Poster Presentation on "A Bio-Inspired Cooperative Algorithm for Dist...
EMBC'13 Poster Presentation on "A Bio-Inspired Cooperative Algorithm for Dist...
Md Kafiul Islam
 
Vision systems_Image processing tool box in MATLAB
Vision systems_Image processing tool box in MATLABVision systems_Image processing tool box in MATLAB
Vision systems_Image processing tool box in MATLAB
Hinna Nayab
 
CenterForDomainSpecificComputing-Poster
CenterForDomainSpecificComputing-PosterCenterForDomainSpecificComputing-Poster
CenterForDomainSpecificComputing-PosterYunming Zhang
 
Iterative Closest Point Algorithm - analysis and implementation
Iterative Closest Point Algorithm - analysis and implementationIterative Closest Point Algorithm - analysis and implementation
Iterative Closest Point Algorithm - analysis and implementation
Pankaj Gautam
 
Distance measuring Algrothim
Distance measuring AlgrothimDistance measuring Algrothim
Distance measuring AlgrothimMuhammad Ahmed
 
Deep Learning A-Z™: Artificial Neural Networks (ANN) - Backpropagation
Deep Learning A-Z™: Artificial Neural Networks (ANN) - BackpropagationDeep Learning A-Z™: Artificial Neural Networks (ANN) - Backpropagation
Deep Learning A-Z™: Artificial Neural Networks (ANN) - Backpropagation
Kirill Eremenko
 
Scale invariant feature transform
Scale invariant feature transformScale invariant feature transform
Scale invariant feature transform
Mohammad Asghar Barech
 
Levenberg - Marquardt (LM) algorithm_ aghazade
Levenberg - Marquardt (LM) algorithm_ aghazadeLevenberg - Marquardt (LM) algorithm_ aghazade
Levenberg - Marquardt (LM) algorithm_ aghazade
Inistute of Geophysics, Tehran university , Tehran/ iran
 
J. Park, AAAI 2022, MLILAB, KAIST AI
J. Park, AAAI 2022, MLILAB, KAIST AIJ. Park, AAAI 2022, MLILAB, KAIST AI
J. Park, AAAI 2022, MLILAB, KAIST AI
MLILAB
 
K Nearest Neighbor Algorithm
K Nearest Neighbor AlgorithmK Nearest Neighbor Algorithm
K Nearest Neighbor Algorithm
Tharuka Vishwajith Sarathchandra
 

What's hot (20)

G. Park, J.-Y. Yang, et. al., NeurIPS 2020, MLILAB, KAIST AI
G. Park, J.-Y. Yang, et. al., NeurIPS 2020, MLILAB, KAIST AIG. Park, J.-Y. Yang, et. al., NeurIPS 2020, MLILAB, KAIST AI
G. Park, J.-Y. Yang, et. al., NeurIPS 2020, MLILAB, KAIST AI
 
J. Park, H. Shim, AAAI 2022, MLILAB, KAISTAI
J. Park, H. Shim, AAAI 2022, MLILAB, KAISTAIJ. Park, H. Shim, AAAI 2022, MLILAB, KAISTAI
J. Park, H. Shim, AAAI 2022, MLILAB, KAISTAI
 
Image Compression
Image CompressionImage Compression
Image Compression
 
IMAGE ENHANCEMENT /ENHANCHING the feature of an image/ Image editor
IMAGE ENHANCEMENT /ENHANCHING the feature of an image/ Image editorIMAGE ENHANCEMENT /ENHANCHING the feature of an image/ Image editor
IMAGE ENHANCEMENT /ENHANCHING the feature of an image/ Image editor
 
Scale Invariant Feature Transform
Scale Invariant Feature TransformScale Invariant Feature Transform
Scale Invariant Feature Transform
 
Hog
HogHog
Hog
 
The Error Bound on SLLN
The Error Bound on SLLNThe Error Bound on SLLN
The Error Bound on SLLN
 
The Error Bound on SLLN
The Error Bound on SLLNThe Error Bound on SLLN
The Error Bound on SLLN
 
EMBC'13 Poster Presentation on "A Bio-Inspired Cooperative Algorithm for Dist...
EMBC'13 Poster Presentation on "A Bio-Inspired Cooperative Algorithm for Dist...EMBC'13 Poster Presentation on "A Bio-Inspired Cooperative Algorithm for Dist...
EMBC'13 Poster Presentation on "A Bio-Inspired Cooperative Algorithm for Dist...
 
Vision systems_Image processing tool box in MATLAB
Vision systems_Image processing tool box in MATLABVision systems_Image processing tool box in MATLAB
Vision systems_Image processing tool box in MATLAB
 
CenterForDomainSpecificComputing-Poster
CenterForDomainSpecificComputing-PosterCenterForDomainSpecificComputing-Poster
CenterForDomainSpecificComputing-Poster
 
Iterative Closest Point Algorithm - analysis and implementation
Iterative Closest Point Algorithm - analysis and implementationIterative Closest Point Algorithm - analysis and implementation
Iterative Closest Point Algorithm - analysis and implementation
 
Distance measuring Algrothim
Distance measuring AlgrothimDistance measuring Algrothim
Distance measuring Algrothim
 
Deep Learning A-Z™: Artificial Neural Networks (ANN) - Backpropagation
Deep Learning A-Z™: Artificial Neural Networks (ANN) - BackpropagationDeep Learning A-Z™: Artificial Neural Networks (ANN) - Backpropagation
Deep Learning A-Z™: Artificial Neural Networks (ANN) - Backpropagation
 
Poster
PosterPoster
Poster
 
Scale invariant feature transform
Scale invariant feature transformScale invariant feature transform
Scale invariant feature transform
 
Levenberg - Marquardt (LM) algorithm_ aghazade
Levenberg - Marquardt (LM) algorithm_ aghazadeLevenberg - Marquardt (LM) algorithm_ aghazade
Levenberg - Marquardt (LM) algorithm_ aghazade
 
J. Park, AAAI 2022, MLILAB, KAIST AI
J. Park, AAAI 2022, MLILAB, KAIST AIJ. Park, AAAI 2022, MLILAB, KAIST AI
J. Park, AAAI 2022, MLILAB, KAIST AI
 
K Nearest Neighbor Algorithm
K Nearest Neighbor AlgorithmK Nearest Neighbor Algorithm
K Nearest Neighbor Algorithm
 
SIFT
SIFTSIFT
SIFT
 

Similar to A Fast Content-Based Image Retrieval Method Using Deep Visual Features

Machine Learning Essentials Demystified part2 | Big Data Demystified
Machine Learning Essentials Demystified part2 | Big Data DemystifiedMachine Learning Essentials Demystified part2 | Big Data Demystified
Machine Learning Essentials Demystified part2 | Big Data Demystified
Omid Vahdaty
 
Salient KeypointSelection for Object Representation
Salient KeypointSelection for Object RepresentationSalient KeypointSelection for Object Representation
Salient KeypointSelection for Object Representation
Prerana Mukherjee
 
Enhancing the performance of kmeans algorithm
Enhancing the performance of kmeans algorithmEnhancing the performance of kmeans algorithm
Enhancing the performance of kmeans algorithm
Hadi Fadlallah
 
Deep Learning Models for Question Answering
Deep Learning Models for Question AnsweringDeep Learning Models for Question Answering
Deep Learning Models for Question Answering
Sujit Pal
 
Experiments in genetic programming
Experiments in genetic programmingExperiments in genetic programming
Experiments in genetic programming
Lars Marius Garshol
 
SECURE IMAGE RETRIEVAL BASED ON HYBRID FEATURES AND HASHES
SECURE IMAGE RETRIEVAL BASED ON HYBRID FEATURES AND HASHESSECURE IMAGE RETRIEVAL BASED ON HYBRID FEATURES AND HASHES
SECURE IMAGE RETRIEVAL BASED ON HYBRID FEATURES AND HASHES
ranjit banshpal
 
MNIST 10-class Classifiers
MNIST 10-class ClassifiersMNIST 10-class Classifiers
MNIST 10-class Classifiers
Sheetal Gangakhedkar
 
Thought Vectors and Knowledge Graphs in AI-powered Search
Thought Vectors and Knowledge Graphs in AI-powered SearchThought Vectors and Knowledge Graphs in AI-powered Search
Thought Vectors and Knowledge Graphs in AI-powered Search
Trey Grainger
 
Implementation of high performance feature extraction method using oriented f...
Implementation of high performance feature extraction method using oriented f...Implementation of high performance feature extraction method using oriented f...
Implementation of high performance feature extraction method using oriented f...
eSAT Journals
 
Practical Data Science: Data Modelling and Presentation
Practical Data Science: Data Modelling and PresentationPractical Data Science: Data Modelling and Presentation
Practical Data Science: Data Modelling and Presentation
HariniMS1
 
Rapid object detection using boosted cascade of simple features
Rapid object detection using boosted  cascade of simple featuresRapid object detection using boosted  cascade of simple features
Rapid object detection using boosted cascade of simple featuresHirantha Pradeep
 
Machine Learning Pipelines
Machine Learning PipelinesMachine Learning Pipelines
Machine Learning Pipelines
jeykottalam
 
Object recognition
Object recognitionObject recognition
Object recognition
saniacorreya
 
Visual diagnostics for more effective machine learning
Visual diagnostics for more effective machine learningVisual diagnostics for more effective machine learning
Visual diagnostics for more effective machine learning
Benjamin Bengfort
 
Rui Meng - 2017 - Deep Keyphrase Generation
Rui Meng - 2017 - Deep Keyphrase GenerationRui Meng - 2017 - Deep Keyphrase Generation
Rui Meng - 2017 - Deep Keyphrase Generation
Association for Computational Linguistics
 
Saliency-based Models of Image Content and their Application to Auto-Annotati...
Saliency-based Models of Image Content and their Application to Auto-Annotati...Saliency-based Models of Image Content and their Application to Auto-Annotati...
Saliency-based Models of Image Content and their Application to Auto-Annotati...
Jonathon Hare
 
Statistics project2
Statistics project2Statistics project2
Statistics project2shri1984
 
Machine Learning, Deep Learning and Data Analysis Introduction
Machine Learning, Deep Learning and Data Analysis IntroductionMachine Learning, Deep Learning and Data Analysis Introduction
Machine Learning, Deep Learning and Data Analysis Introduction
Te-Yen Liu
 
Clustering Methods with R
Clustering Methods with RClustering Methods with R
Clustering Methods with R
Akira Murakami
 
Unleashing the Power of Machine Learning Prototyping Using Azure AutoML and P...
Unleashing the Power of Machine Learning Prototyping Using Azure AutoML and P...Unleashing the Power of Machine Learning Prototyping Using Azure AutoML and P...
Unleashing the Power of Machine Learning Prototyping Using Azure AutoML and P...
Luca Zavarella
 

Similar to A Fast Content-Based Image Retrieval Method Using Deep Visual Features (20)

Machine Learning Essentials Demystified part2 | Big Data Demystified
Machine Learning Essentials Demystified part2 | Big Data DemystifiedMachine Learning Essentials Demystified part2 | Big Data Demystified
Machine Learning Essentials Demystified part2 | Big Data Demystified
 
Salient KeypointSelection for Object Representation
Salient KeypointSelection for Object RepresentationSalient KeypointSelection for Object Representation
Salient KeypointSelection for Object Representation
 
Enhancing the performance of kmeans algorithm
Enhancing the performance of kmeans algorithmEnhancing the performance of kmeans algorithm
Enhancing the performance of kmeans algorithm
 
Deep Learning Models for Question Answering
Deep Learning Models for Question AnsweringDeep Learning Models for Question Answering
Deep Learning Models for Question Answering
 
Experiments in genetic programming
Experiments in genetic programmingExperiments in genetic programming
Experiments in genetic programming
 
SECURE IMAGE RETRIEVAL BASED ON HYBRID FEATURES AND HASHES
SECURE IMAGE RETRIEVAL BASED ON HYBRID FEATURES AND HASHESSECURE IMAGE RETRIEVAL BASED ON HYBRID FEATURES AND HASHES
SECURE IMAGE RETRIEVAL BASED ON HYBRID FEATURES AND HASHES
 
MNIST 10-class Classifiers
MNIST 10-class ClassifiersMNIST 10-class Classifiers
MNIST 10-class Classifiers
 
Thought Vectors and Knowledge Graphs in AI-powered Search
Thought Vectors and Knowledge Graphs in AI-powered SearchThought Vectors and Knowledge Graphs in AI-powered Search
Thought Vectors and Knowledge Graphs in AI-powered Search
 
Implementation of high performance feature extraction method using oriented f...
Implementation of high performance feature extraction method using oriented f...Implementation of high performance feature extraction method using oriented f...
Implementation of high performance feature extraction method using oriented f...
 
Practical Data Science: Data Modelling and Presentation
Practical Data Science: Data Modelling and PresentationPractical Data Science: Data Modelling and Presentation
Practical Data Science: Data Modelling and Presentation
 
Rapid object detection using boosted cascade of simple features
Rapid object detection using boosted  cascade of simple featuresRapid object detection using boosted  cascade of simple features
Rapid object detection using boosted cascade of simple features
 
Machine Learning Pipelines
Machine Learning PipelinesMachine Learning Pipelines
Machine Learning Pipelines
 
Object recognition
Object recognitionObject recognition
Object recognition
 
Visual diagnostics for more effective machine learning
Visual diagnostics for more effective machine learningVisual diagnostics for more effective machine learning
Visual diagnostics for more effective machine learning
 
Rui Meng - 2017 - Deep Keyphrase Generation
Rui Meng - 2017 - Deep Keyphrase GenerationRui Meng - 2017 - Deep Keyphrase Generation
Rui Meng - 2017 - Deep Keyphrase Generation
 
Saliency-based Models of Image Content and their Application to Auto-Annotati...
Saliency-based Models of Image Content and their Application to Auto-Annotati...Saliency-based Models of Image Content and their Application to Auto-Annotati...
Saliency-based Models of Image Content and their Application to Auto-Annotati...
 
Statistics project2
Statistics project2Statistics project2
Statistics project2
 
Machine Learning, Deep Learning and Data Analysis Introduction
Machine Learning, Deep Learning and Data Analysis IntroductionMachine Learning, Deep Learning and Data Analysis Introduction
Machine Learning, Deep Learning and Data Analysis Introduction
 
Clustering Methods with R
Clustering Methods with RClustering Methods with R
Clustering Methods with R
 
Unleashing the Power of Machine Learning Prototyping Using Azure AutoML and P...
Unleashing the Power of Machine Learning Prototyping Using Azure AutoML and P...Unleashing the Power of Machine Learning Prototyping Using Azure AutoML and P...
Unleashing the Power of Machine Learning Prototyping Using Azure AutoML and P...
 

More from Hiroki Tanioka

PoseNet(ml5.js)を用いた投球フォーム推定
PoseNet(ml5.js)を用いた投球フォーム推定PoseNet(ml5.js)を用いた投球フォーム推定
PoseNet(ml5.js)を用いた投球フォーム推定
Hiroki Tanioka
 
おとなのプログラミング教室 vol1
おとなのプログラミング教室 vol1おとなのプログラミング教室 vol1
おとなのプログラミング教室 vol1
Hiroki Tanioka
 
Mentoring without Technical Skills
Mentoring without Technical SkillsMentoring without Technical Skills
Mentoring without Technical Skills
Hiroki Tanioka
 
グループ学習で学ぶプログラミング 〜さあ、いっしょに考えよう!〜
グループ学習で学ぶプログラミング 〜さあ、いっしょに考えよう!〜グループ学習で学ぶプログラミング 〜さあ、いっしょに考えよう!〜
グループ学習で学ぶプログラミング 〜さあ、いっしょに考えよう!〜
Hiroki Tanioka
 
理想のAIと現実の機械学習
理想のAIと現実の機械学習理想のAIと現実の機械学習
理想のAIと現実の機械学習
Hiroki Tanioka
 
Super Easy Way of Building Image Search with Keras
Super Easy Way of Building Image Search with KerasSuper Easy Way of Building Image Search with Keras
Super Easy Way of Building Image Search with Keras
Hiroki Tanioka
 
ソフトウェア開発の心得
ソフトウェア開発の心得ソフトウェア開発の心得
ソフトウェア開発の心得Hiroki Tanioka
 

More from Hiroki Tanioka (7)

PoseNet(ml5.js)を用いた投球フォーム推定
PoseNet(ml5.js)を用いた投球フォーム推定PoseNet(ml5.js)を用いた投球フォーム推定
PoseNet(ml5.js)を用いた投球フォーム推定
 
おとなのプログラミング教室 vol1
おとなのプログラミング教室 vol1おとなのプログラミング教室 vol1
おとなのプログラミング教室 vol1
 
Mentoring without Technical Skills
Mentoring without Technical SkillsMentoring without Technical Skills
Mentoring without Technical Skills
 
グループ学習で学ぶプログラミング 〜さあ、いっしょに考えよう!〜
グループ学習で学ぶプログラミング 〜さあ、いっしょに考えよう!〜グループ学習で学ぶプログラミング 〜さあ、いっしょに考えよう!〜
グループ学習で学ぶプログラミング 〜さあ、いっしょに考えよう!〜
 
理想のAIと現実の機械学習
理想のAIと現実の機械学習理想のAIと現実の機械学習
理想のAIと現実の機械学習
 
Super Easy Way of Building Image Search with Keras
Super Easy Way of Building Image Search with KerasSuper Easy Way of Building Image Search with Keras
Super Easy Way of Building Image Search with Keras
 
ソフトウェア開発の心得
ソフトウェア開発の心得ソフトウェア開発の心得
ソフトウェア開発の心得
 

Recently uploaded

ASME IX(9) 2007 Full Version .pdf
ASME IX(9)  2007 Full Version       .pdfASME IX(9)  2007 Full Version       .pdf
ASME IX(9) 2007 Full Version .pdf
AhmedHussein950959
 
ethical hacking in wireless-hacking1.ppt
ethical hacking in wireless-hacking1.pptethical hacking in wireless-hacking1.ppt
ethical hacking in wireless-hacking1.ppt
Jayaprasanna4
 
Water Industry Process Automation and Control Monthly - May 2024.pdf
Water Industry Process Automation and Control Monthly - May 2024.pdfWater Industry Process Automation and Control Monthly - May 2024.pdf
Water Industry Process Automation and Control Monthly - May 2024.pdf
Water Industry Process Automation & Control
 
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
zwunae
 
The Benefits and Techniques of Trenchless Pipe Repair.pdf
The Benefits and Techniques of Trenchless Pipe Repair.pdfThe Benefits and Techniques of Trenchless Pipe Repair.pdf
The Benefits and Techniques of Trenchless Pipe Repair.pdf
Pipe Restoration Solutions
 
HYDROPOWER - Hydroelectric power generation
HYDROPOWER - Hydroelectric power generationHYDROPOWER - Hydroelectric power generation
HYDROPOWER - Hydroelectric power generation
Robbie Edward Sayers
 
Design and Analysis of Algorithms-DP,Backtracking,Graphs,B&B
Design and Analysis of Algorithms-DP,Backtracking,Graphs,B&BDesign and Analysis of Algorithms-DP,Backtracking,Graphs,B&B
Design and Analysis of Algorithms-DP,Backtracking,Graphs,B&B
Sreedhar Chowdam
 
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
ydteq
 
ML for identifying fraud using open blockchain data.pptx
ML for identifying fraud using open blockchain data.pptxML for identifying fraud using open blockchain data.pptx
ML for identifying fraud using open blockchain data.pptx
Vijay Dialani, PhD
 
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
bakpo1
 
AP LAB PPT.pdf ap lab ppt no title specific
AP LAB PPT.pdf ap lab ppt no title specificAP LAB PPT.pdf ap lab ppt no title specific
AP LAB PPT.pdf ap lab ppt no title specific
BrazilAccount1
 
Student information management system project report ii.pdf
Student information management system project report ii.pdfStudent information management system project report ii.pdf
Student information management system project report ii.pdf
Kamal Acharya
 
RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...
RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...
RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...
thanhdowork
 
DESIGN A COTTON SEED SEPARATION MACHINE.docx
DESIGN A COTTON SEED SEPARATION MACHINE.docxDESIGN A COTTON SEED SEPARATION MACHINE.docx
DESIGN A COTTON SEED SEPARATION MACHINE.docx
FluxPrime1
 
power quality voltage fluctuation UNIT - I.pptx
power quality voltage fluctuation UNIT - I.pptxpower quality voltage fluctuation UNIT - I.pptx
power quality voltage fluctuation UNIT - I.pptx
ViniHema
 
Runway Orientation Based on the Wind Rose Diagram.pptx
Runway Orientation Based on the Wind Rose Diagram.pptxRunway Orientation Based on the Wind Rose Diagram.pptx
Runway Orientation Based on the Wind Rose Diagram.pptx
SupreethSP4
 
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdf
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdfHybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdf
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdf
fxintegritypublishin
 
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...
Dr.Costas Sachpazis
 
在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样
在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样
在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样
obonagu
 
Gen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdfGen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdf
gdsczhcet
 

Recently uploaded (20)

ASME IX(9) 2007 Full Version .pdf
ASME IX(9)  2007 Full Version       .pdfASME IX(9)  2007 Full Version       .pdf
ASME IX(9) 2007 Full Version .pdf
 
ethical hacking in wireless-hacking1.ppt
ethical hacking in wireless-hacking1.pptethical hacking in wireless-hacking1.ppt
ethical hacking in wireless-hacking1.ppt
 
Water Industry Process Automation and Control Monthly - May 2024.pdf
Water Industry Process Automation and Control Monthly - May 2024.pdfWater Industry Process Automation and Control Monthly - May 2024.pdf
Water Industry Process Automation and Control Monthly - May 2024.pdf
 
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
 
The Benefits and Techniques of Trenchless Pipe Repair.pdf
The Benefits and Techniques of Trenchless Pipe Repair.pdfThe Benefits and Techniques of Trenchless Pipe Repair.pdf
The Benefits and Techniques of Trenchless Pipe Repair.pdf
 
HYDROPOWER - Hydroelectric power generation
HYDROPOWER - Hydroelectric power generationHYDROPOWER - Hydroelectric power generation
HYDROPOWER - Hydroelectric power generation
 
Design and Analysis of Algorithms-DP,Backtracking,Graphs,B&B
Design and Analysis of Algorithms-DP,Backtracking,Graphs,B&BDesign and Analysis of Algorithms-DP,Backtracking,Graphs,B&B
Design and Analysis of Algorithms-DP,Backtracking,Graphs,B&B
 
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
 
ML for identifying fraud using open blockchain data.pptx
ML for identifying fraud using open blockchain data.pptxML for identifying fraud using open blockchain data.pptx
ML for identifying fraud using open blockchain data.pptx
 
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
 
AP LAB PPT.pdf ap lab ppt no title specific
AP LAB PPT.pdf ap lab ppt no title specificAP LAB PPT.pdf ap lab ppt no title specific
AP LAB PPT.pdf ap lab ppt no title specific
 
Student information management system project report ii.pdf
Student information management system project report ii.pdfStudent information management system project report ii.pdf
Student information management system project report ii.pdf
 
RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...
RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...
RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...
 
DESIGN A COTTON SEED SEPARATION MACHINE.docx
DESIGN A COTTON SEED SEPARATION MACHINE.docxDESIGN A COTTON SEED SEPARATION MACHINE.docx
DESIGN A COTTON SEED SEPARATION MACHINE.docx
 
power quality voltage fluctuation UNIT - I.pptx
power quality voltage fluctuation UNIT - I.pptxpower quality voltage fluctuation UNIT - I.pptx
power quality voltage fluctuation UNIT - I.pptx
 
Runway Orientation Based on the Wind Rose Diagram.pptx
Runway Orientation Based on the Wind Rose Diagram.pptxRunway Orientation Based on the Wind Rose Diagram.pptx
Runway Orientation Based on the Wind Rose Diagram.pptx
 
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdf
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdfHybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdf
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdf
 
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...
 
在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样
在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样
在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样
 
Gen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdfGen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdf
 

A Fast Content-Based Image Retrieval Method Using Deep Visual Features

  • 1. A Fast Content-Based Image Retrieval Method Using Deep Visual Features Hiroki Tanioka (Assist. Prof. Ph.D.) Tokushima University, Japan
  • 2. HELLO! I am Hiroki Tanioka. You can find me at @taniokah 2
  • 3. “Quotations are why we cannot use cosine similarity on inverted index-based search engine, despite cosine similarity is commonly used for many classification problems.” 3
  • 4. Scoring Method Includes normalizing every dot product score with L2 norm of each indexed document. THERE ARE TWO REASONS FOR THAT Higher Dimension Needs much cost of calculating dot product between query vector and indexed document vector. 4 Calculation Faster Dimension Lower
  • 5. Small Ideas My key concept is using normalized index and truncating feature vectors. 5
  • 6. 1. FASTER COSINE SMILARITY Let’s start making index with normalized vectors in advance.
  • 8. 2. LOWER VECTOR DIMENSTION Let’s reduce features un-preferable document features for score.
  • 9. SORT AND REDUCE FEATURE VECTOR 9 ・ ・ ・
  • 10. Dataset Statistics Number of Images: 25,000 Type: Dogs and Cats Source: Kaggle Search Engine Elasticsearch 7.0.1 INFROMATION OF EXPERIMENTAL ENVELONMENT Image Features Model: VGG-16 Number of Features: 1,000 Score: Probability Machine Specification MacBook Pro, 2.5 GHz, 16GB (Retina, 15-inch, Mid 2015) 10
  • 11. THREE TEST PATERNS FOR EVALUATION OF MY SYSTEM Finding a picture with original picture Finding a picture with low-resolution Finding a picture with part of picture 11
  • 12. AVERAGE RESPONSE TIME Feature 1 2 3 4 5 10 20 30 40 50 100 200 400 1000 dot-prod 0.23 0.33 0.25 0.36 0.37 0.42 0.74 0.69 0.80 0.86 1.43 2.02 2.86 2.87 L1 0.27 0.27 0.30 0.31 0.35 0.35 0.57 0.80 0.70 0.72 1.35 1.58 2.58 2.85 L2 0.26 0.28 0.28 0.30 0.29 0.35 0.51 0.63 0.80 0.99 1.97 3.80 5.78 6.56 cosine 0.23 0.25 0.27 0.28 0.30 0.35 0.64 0.62 0.61 0.59 1.14 1.43 2.60 3.08 dot+cos 44.0 42.2 39.7 36.1 35.0 36.1 35.4 35.0 37.9 41.2 47.0 46.5 52.0 53.0 12 FAST ENOUGH
  • 13. 13 Comparison MAP of feature number in 100 top-k results and 1.0 resolution rate. GOOD ENOUGH
  • 14. 14 Comparison MAP of resolution rates, in 1,000 feature number and 100 top-k results. FAIR ENOUGH
  • 15. 15 Comparison MAP of feature numbers using quartered partial image query, in 1,000 feature number and 100 top-k results. FAVORABLE
  • 16. Pre-Normalizing Is effective for calculating cosine similarity faster under 100 features. THERE ARE TWO RESULTS Reducing Features Is useful, but at least 8 features are required for our image search experiment. 16
  • 17. INSTRUCTIONS FOR USE USE XPACK IN ELASTICSEARCH https://www.elastic.co/blog/text- similarity-search-with-vectors-in- elasticsearch Elasticsearch 7.3 released cosine_similarity in xpack. Still, L1 and L2 are under construction. VISIT TO MY GITHUB PAGE https://github.com/taniokah/icdar- wml-2019 I’ve just already left my messy scripts on Github. I will explain an instruction for my project. 17
  • 18. 18 THANKS! Any questions? You can find me at: ❑ @taniokah ❑ tanioka.hiroki@tokushima-u.ac.jp This work was supported by JSPS KAKENHI Grant Number JP18H03344.

Editor's Notes

  1. Deep Convolutional representations Deep Visual Features Dot Product and TF-IDF
  2. Exact image search Find reused images Detect copyright violation 10 images for test 25,000 images for index