SlideShare a Scribd company logo
An accurate retrieval through R-MAC+
descriptors for landmark recognition
Federico Magliani, Andrea Prati
ICDSC 2018 – Eindhoven, Netherlands – 3-4 September 2018
Agenda
2
➢ Motivations
➢ Summary of contributions
➢ Related works
➢ Introduction to R-MAC descriptors
➢ Proposed approach (R-MAC+)
➢ Experimental results
➢ Conclusions
Motivations
3
Landmark Recognition problem
➢ Try to understand what’s is in front
of you and retrieve similar images.
➢ Semantic gap: for a human, this task
is pretty simple thanks to personal
experience, but a computer can use
only the info available in the images.
➢ It is far from being solved
(viewpoint, illumination conditions,
image resolution, ...).
Motivations
4
➢ Challenges
○ High accuracy retrieval (precision)
○ Fast research (response to query)
○ Reduced memory occupied (mobile friendly)
○ Work well with big data (>1M data)
➢ Possible applications
○ Augmented reality (tourism)
○ Person Re-ID (video-surveillance)
○ Online clothes search (fashion)
Agenda
5
➢ Motivations
➢ Summary of contributions
➢ Related works
➢ Introduction to R-MAC descriptors
➢ Proposed approach (R-MAC+)
➢ Experimental results
➢ Conclusions and Future Works
Summary of contributions
6
➢ a new region detector for CNN feature maps implemented through grids, that respect
the aspect ratio of the images.
➢ an improvement on the effectiveness of the multi-resolution approach for R-MAC
descriptors.
➢ a novel retrieval method for checking the similarities between query descriptors and
regions of database R-MAC descriptors. It allows to outperform the results of R-MAC
descriptors on Oxford5k and Paris6k by +7% and +3%.
Agenda
7
➢ Motivations
➢ Summary of contribution
➢ Related works
➢ Introduction to R-MAC descriptors
➢ Proposed approach (R-MAC+)
➢ Experimental results
➢ Conclusions
Related works
8
➢ Bag of Words (BoW): first method for solving the problem (different
techniques: vocabulary tree, …).
➢ VLAD: similar to BoW, but using the residual of the descriptors
(=feature descriptor - closest centers in the vocabulary).
➢ CNN based: extract features from intermediate layers of CNN
architectures and then apply previous embedding techniques (BLCF, ...).
➢ MAC: max pooling applied on CNN features
➢ R-MAC: regional MAC descriptors created through the application of a
rigid-grid mechanism
Agenda
9
➢ Motivations
➢ Summary of contributions
➢ Related works
➢ Introduction to R-MAC descriptors
➢ Proposed approach (R-MAC+)
➢ Experimental results
➢ Conclusions
R-MAC (Regional MAC) descriptors
10
Considering a rectangular region R ⊆ Ω = (1,W) x (1,H), and define the regional feature vector:
fR
= (fR,1
...fR,i
...fR,K
)T
where fR,i
= max Xi
(p) is the maximum activation of the ith
channel on the considered
region.
Then we calculate the feature vector associated with each region, and post-process it with
l2
-normalization, PCA-whitening and l2
-normalization. We combine the collection of regional feature
vectors into a single image vector by summing them and l2
-normalizing in the end.
We define the response maps and sample square regions at
L different scales
➢ at the largest scale (l=1), the region size is determined
to be as large as possible (height = width = min(W,H))
➢ at every other scale l, we uniformly sample l x (l+m-1)
regions of width 2min(W,H)/(l+1). (with m=2)
R-MAC (Regional MAC) descriptors
11
Settings:
➢ Fully convolutional off-the-shelf VGG16
➢ Pool5
➢ Spatial Max pooling
➢ High Resolution images
➢ Global descriptor based on aggregating region vectors
➢ Sliding window approach
Tolias et al. Particular object retrieval with integral max-pooling of CNN activations. arXiv 2015.
Agenda
12
➢ Motivations
➢ Summary of contributions
➢ Related works
➢ Introduction to R-MAC descriptors
➢ Proposed approach (R-MAC+)
➢ Experimental results
➢ Conclusions
Proposed approach: R-MAC+
New multi-resolution approach: the images are resized of +25%,-25%, 0% on the largest
size, respecting the aspect ratio of the image.
➢ This strategy is an alternative of the first multi-resolution approach, that resized the
image to a fixed size: 550px, 800px and 1050 on the largest size, retaining the aspect
ratio of the image.
➢ This strategy should allow to augment the dimensions of the feature maps in order to
have more features and therefore local maxima than the previous multi-resolution
R-MAC. This approach is connected to the new region detector, that detects a
reduced number of regions (15) instead of the 20 of the original one.
13
Proposed approach: R-MAC+
14
A new mechanism for region detection in the CNN feature maps (15 regions)
● l=0 → 1 region covering entirely the image;
● l=1 → 2 square regions (widthRegion = heightRegion = min(H,W));
● l=2 → 6 rect regions (widthRegion = heightRegion =⌈2*min(W,H)/(l+1))⌉, arranged along the
horizontal axis (width and height of the regions are adapted to cover all the image);
● l=3 → 6 rect regions (widthRegion = heightRegion= ⌈2*min(W,H)/(l+2))⌉, arranged along the
vertical axis (width and height of the regions are adapted to cover all the image).
Proposed approach: R-MAC+
15
A new retrieval method based on db regions (MAC descriptors of the database images) and the
R-MAC descriptors of the query images (+7% on Oxford5k and +4% on Paris6k than previous results)
Agenda
16
➢ Motivations
➢ Summary of contributions
➢ Related works
➢ Introduction to R-MAC descriptors
➢ Proposed approach (R-MAC+)
➢ Experimental results
➢ Conclusions
Datasets and evaluation metric
Datasets:
➢ Holidays (1491 images: 500 classes, 500 queries).
➢ Oxford5k (5063 images, 11 classes, 55 queries).
➢ Paris6k (6412 images, 11 classes, 55 queries).
Evaluation metric:
➢ mAP (mean Average Precision) → mean of Average Precision scores (correct results)
for each query, based on the position in the ranking.
17
Results
18
Method Network Holidays
(original/rotated)
Oxf5k Paris6k
MAC VGG19 76.26 % 57.44 % 73.15 %
R-MAC VGG19 87.65 % 65.56 % 82.80 %
R-MAC ResNet50 92.55 % 71.77 % 83.31 %
M-R R-MAC+ ResNet50 94.63 % / 95.58 % 78.88 % 88.63 %
M-R R-MAC+ with retrieval
based on db regions
ResNet50 94.37 % / 95.87 % 85.39 % 91.90 %
Results after QE application
19
Method Network Holidays
(original/rotated)
Oxf5k Paris6k
M-R R-MAC+ ResNet50 94.97 % / 95.97 % 86.45 % 92.01 %
M-R R-MAC+ with retrieval
based on db regions
ResNet50 94.42 % / 96.05 % 87.92 % 93.64 %
M-R R-MAC+ with retrieval
based on db regions and query
expansion based on db regions
ResNet50 94.28 % / 95.91 % 88.78 % 92.30 %
Comparison with the state of the art
20
Agenda
21
➢ Motivations
➢ Summary of contributions
➢ Related works
➢ Introduction to R-MAC descriptors
➢ Proposed approach (R-MAC+)
➢ Experimental results
➢ Conclusions
Conclusions
➢ We propose different improvements on R-MAC descriptors in order to make the
retrieval very accurate.
○ A multi-resolution approach, that uses bigger feature maps than the previous one.
○ A new region detector with the use of adaptable grids allows to catch more local
maxima.
○ A novel retrieval method based on db regions that highly boosts the performance on
Oxford5k and Paris6k.
➢ The proposed method outperforms the state of the art on Holidays, both on the
original and rotated version. Also it outperforms the state-of-the-art results on
some other public benchmarks without the fine-tuning application.
22
Thank you for your attention!
questions?
http://implab.ce.unipr.it
23

More Related Content

What's hot

Densebox
DenseboxDensebox
Densebox
冠宇 陳
 
Aerial detection part3
Aerial detection part3Aerial detection part3
Aerial detection part3
ssuser456ad6
 
Feature pyramid networks for object detection
Feature pyramid networks for object detection Feature pyramid networks for object detection
Feature pyramid networks for object detection
heedaeKwon
 
Automatic road environment classification 20121002
Automatic road environment classification 20121002Automatic road environment classification 20121002
Automatic road environment classification 20121002
es712
 
NetVLAD: CNN architecture for weakly supervised place recognition
NetVLAD:  CNN architecture for weakly supervised place recognitionNetVLAD:  CNN architecture for weakly supervised place recognition
NetVLAD: CNN architecture for weakly supervised place recognition
Geunhee Cho
 
Path Planning And Navigation
Path Planning And NavigationPath Planning And Navigation
Path Planning And Navigation
guest90654fd
 
06466595
0646659506466595
06466595
Jéssyca Bessa
 
Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013
Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013
Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013
Sunando Sengupta
 
K-BestMatch
K-BestMatchK-BestMatch
K-BestMatch
Roberto Trasarti
 
object detection paper review
object detection paper reviewobject detection paper review
object detection paper review
Yoonho Na
 
presentazione_IGARSS2011.ppt
presentazione_IGARSS2011.pptpresentazione_IGARSS2011.ppt
presentazione_IGARSS2011.ppt
grssieee
 
mid_presentation
mid_presentationmid_presentation
mid_presentation
Chiraz Nafouki
 
computervision project
computervision projectcomputervision project
computervision project
Lianli Liu
 
Implementation of a lane-tracking system for autonomous driving using Kalman ...
Implementation of a lane-tracking system for autonomous driving using Kalman ...Implementation of a lane-tracking system for autonomous driving using Kalman ...
Implementation of a lane-tracking system for autonomous driving using Kalman ...
Francesco Corucci
 
Automatic Dense Semantic Mapping From Visual Street-level Imagery
Automatic Dense Semantic Mapping From Visual Street-level ImageryAutomatic Dense Semantic Mapping From Visual Street-level Imagery
Automatic Dense Semantic Mapping From Visual Street-level Imagery
Sunando Sengupta
 
Report bep thomas_blanken
Report bep thomas_blankenReport bep thomas_blanken
Report bep thomas_blanken
xepost
 
MS Project
MS ProjectMS Project
Prunet, Pascal: Plume detection and characterization from XCO2 imagery: Evalu...
Prunet, Pascal: Plume detection and characterization from XCO2 imagery: Evalu...Prunet, Pascal: Plume detection and characterization from XCO2 imagery: Evalu...
Prunet, Pascal: Plume detection and characterization from XCO2 imagery: Evalu...
Integrated Carbon Observation System (ICOS)
 
REVIEW OF LANE DETECTION AND TRACKING ALGORITHMS IN ADVANCED DRIVER ASSISTANC...
REVIEW OF LANE DETECTION AND TRACKING ALGORITHMS IN ADVANCED DRIVER ASSISTANC...REVIEW OF LANE DETECTION AND TRACKING ALGORITHMS IN ADVANCED DRIVER ASSISTANC...
REVIEW OF LANE DETECTION AND TRACKING ALGORITHMS IN ADVANCED DRIVER ASSISTANC...
ijcsit
 
Mmclass5b
Mmclass5bMmclass5b
Mmclass5b
Hassan Dar
 

What's hot (20)

Densebox
DenseboxDensebox
Densebox
 
Aerial detection part3
Aerial detection part3Aerial detection part3
Aerial detection part3
 
Feature pyramid networks for object detection
Feature pyramid networks for object detection Feature pyramid networks for object detection
Feature pyramid networks for object detection
 
Automatic road environment classification 20121002
Automatic road environment classification 20121002Automatic road environment classification 20121002
Automatic road environment classification 20121002
 
NetVLAD: CNN architecture for weakly supervised place recognition
NetVLAD:  CNN architecture for weakly supervised place recognitionNetVLAD:  CNN architecture for weakly supervised place recognition
NetVLAD: CNN architecture for weakly supervised place recognition
 
Path Planning And Navigation
Path Planning And NavigationPath Planning And Navigation
Path Planning And Navigation
 
06466595
0646659506466595
06466595
 
Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013
Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013
Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013
 
K-BestMatch
K-BestMatchK-BestMatch
K-BestMatch
 
object detection paper review
object detection paper reviewobject detection paper review
object detection paper review
 
presentazione_IGARSS2011.ppt
presentazione_IGARSS2011.pptpresentazione_IGARSS2011.ppt
presentazione_IGARSS2011.ppt
 
mid_presentation
mid_presentationmid_presentation
mid_presentation
 
computervision project
computervision projectcomputervision project
computervision project
 
Implementation of a lane-tracking system for autonomous driving using Kalman ...
Implementation of a lane-tracking system for autonomous driving using Kalman ...Implementation of a lane-tracking system for autonomous driving using Kalman ...
Implementation of a lane-tracking system for autonomous driving using Kalman ...
 
Automatic Dense Semantic Mapping From Visual Street-level Imagery
Automatic Dense Semantic Mapping From Visual Street-level ImageryAutomatic Dense Semantic Mapping From Visual Street-level Imagery
Automatic Dense Semantic Mapping From Visual Street-level Imagery
 
Report bep thomas_blanken
Report bep thomas_blankenReport bep thomas_blanken
Report bep thomas_blanken
 
MS Project
MS ProjectMS Project
MS Project
 
Prunet, Pascal: Plume detection and characterization from XCO2 imagery: Evalu...
Prunet, Pascal: Plume detection and characterization from XCO2 imagery: Evalu...Prunet, Pascal: Plume detection and characterization from XCO2 imagery: Evalu...
Prunet, Pascal: Plume detection and characterization from XCO2 imagery: Evalu...
 
REVIEW OF LANE DETECTION AND TRACKING ALGORITHMS IN ADVANCED DRIVER ASSISTANC...
REVIEW OF LANE DETECTION AND TRACKING ALGORITHMS IN ADVANCED DRIVER ASSISTANC...REVIEW OF LANE DETECTION AND TRACKING ALGORITHMS IN ADVANCED DRIVER ASSISTANC...
REVIEW OF LANE DETECTION AND TRACKING ALGORITHMS IN ADVANCED DRIVER ASSISTANC...
 
Mmclass5b
Mmclass5bMmclass5b
Mmclass5b
 

Similar to An accurate retrieval through R-MAC+ descriptors for landmark recognition

Module-5-1_230523_171754 (1).pdf
Module-5-1_230523_171754 (1).pdfModule-5-1_230523_171754 (1).pdf
Module-5-1_230523_171754 (1).pdf
vikasmittal92
 
PCA-SIFT: A More Distinctive Representation for Local Image Descriptors
PCA-SIFT: A More Distinctive Representation for Local Image DescriptorsPCA-SIFT: A More Distinctive Representation for Local Image Descriptors
PCA-SIFT: A More Distinctive Representation for Local Image Descriptors
wolf
 
A hybrid sine cosine optimization algorithm for solving global optimization p...
A hybrid sine cosine optimization algorithm for solving global optimization p...A hybrid sine cosine optimization algorithm for solving global optimization p...
A hybrid sine cosine optimization algorithm for solving global optimization p...
Aboul Ella Hassanien
 
FV_IGARSS11.ppt
FV_IGARSS11.pptFV_IGARSS11.ppt
FV_IGARSS11.ppt
grssieee
 
FV_IGARSS11.ppt
FV_IGARSS11.pptFV_IGARSS11.ppt
FV_IGARSS11.ppt
grssieee
 
FV_IGARSS11.ppt
FV_IGARSS11.pptFV_IGARSS11.ppt
FV_IGARSS11.ppt
grssieee
 
FV_IGARSS11.ppt
FV_IGARSS11.pptFV_IGARSS11.ppt
FV_IGARSS11.ppt
grssieee
 
FR3.L09 - MULTIBASELINE GRADIENT AMBIGUITY RESOLUTION TO SUPPORT MINIMUM COST...
FR3.L09 - MULTIBASELINE GRADIENT AMBIGUITY RESOLUTION TO SUPPORT MINIMUM COST...FR3.L09 - MULTIBASELINE GRADIENT AMBIGUITY RESOLUTION TO SUPPORT MINIMUM COST...
FR3.L09 - MULTIBASELINE GRADIENT AMBIGUITY RESOLUTION TO SUPPORT MINIMUM COST...
grssieee
 
Video Stitching using Improved RANSAC and SIFT
Video Stitching using Improved RANSAC and SIFTVideo Stitching using Improved RANSAC and SIFT
Video Stitching using Improved RANSAC and SIFT
IRJET Journal
 
All projects
All projectsAll projects
All projects
Karishma Jain
 
ICRA Nathan Piasco
ICRA Nathan PiascoICRA Nathan Piasco
ICRA Nathan Piasco
Nathan Piasco
 
Convolutional Neural Network for pixel-wise skyline detection
Convolutional Neural Network for pixel-wise skyline detectionConvolutional Neural Network for pixel-wise skyline detection
Convolutional Neural Network for pixel-wise skyline detection
Darian Frajberg
 
NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
taeseon ryu
 
Real time traffic management - challenges and solutions
Real time traffic management - challenges and solutionsReal time traffic management - challenges and solutions
Real time traffic management - challenges and solutions
Institute for Transport Studies (ITS)
 
Computer Vision: Feature matching with RANSAC Algorithm
Computer Vision: Feature matching with RANSAC AlgorithmComputer Vision: Feature matching with RANSAC Algorithm
Computer Vision: Feature matching with RANSAC Algorithm
allyn joy calcaben
 
crowd counting.pptx
crowd counting.pptxcrowd counting.pptx
crowd counting.pptx
shubhampawar445982
 
Landmark Retrieval & Recognition
Landmark Retrieval & RecognitionLandmark Retrieval & Recognition
Landmark Retrieval & Recognition
kenluck2001
 
Deep image retrieval learning global representations for image search
Deep image retrieval  learning global representations for image searchDeep image retrieval  learning global representations for image search
Deep image retrieval learning global representations for image search
Universitat Politècnica de Catalunya
 
Towards better performance: phase congruency based face recognition
Towards better performance: phase congruency based face recognitionTowards better performance: phase congruency based face recognition
Towards better performance: phase congruency based face recognition
TELKOMNIKA JOURNAL
 
20210226 esa-science-coffee-v2.0
20210226 esa-science-coffee-v2.020210226 esa-science-coffee-v2.0
20210226 esa-science-coffee-v2.0
Advanced-Concepts-Team
 

Similar to An accurate retrieval through R-MAC+ descriptors for landmark recognition (20)

Module-5-1_230523_171754 (1).pdf
Module-5-1_230523_171754 (1).pdfModule-5-1_230523_171754 (1).pdf
Module-5-1_230523_171754 (1).pdf
 
PCA-SIFT: A More Distinctive Representation for Local Image Descriptors
PCA-SIFT: A More Distinctive Representation for Local Image DescriptorsPCA-SIFT: A More Distinctive Representation for Local Image Descriptors
PCA-SIFT: A More Distinctive Representation for Local Image Descriptors
 
A hybrid sine cosine optimization algorithm for solving global optimization p...
A hybrid sine cosine optimization algorithm for solving global optimization p...A hybrid sine cosine optimization algorithm for solving global optimization p...
A hybrid sine cosine optimization algorithm for solving global optimization p...
 
FV_IGARSS11.ppt
FV_IGARSS11.pptFV_IGARSS11.ppt
FV_IGARSS11.ppt
 
FV_IGARSS11.ppt
FV_IGARSS11.pptFV_IGARSS11.ppt
FV_IGARSS11.ppt
 
FV_IGARSS11.ppt
FV_IGARSS11.pptFV_IGARSS11.ppt
FV_IGARSS11.ppt
 
FV_IGARSS11.ppt
FV_IGARSS11.pptFV_IGARSS11.ppt
FV_IGARSS11.ppt
 
FR3.L09 - MULTIBASELINE GRADIENT AMBIGUITY RESOLUTION TO SUPPORT MINIMUM COST...
FR3.L09 - MULTIBASELINE GRADIENT AMBIGUITY RESOLUTION TO SUPPORT MINIMUM COST...FR3.L09 - MULTIBASELINE GRADIENT AMBIGUITY RESOLUTION TO SUPPORT MINIMUM COST...
FR3.L09 - MULTIBASELINE GRADIENT AMBIGUITY RESOLUTION TO SUPPORT MINIMUM COST...
 
Video Stitching using Improved RANSAC and SIFT
Video Stitching using Improved RANSAC and SIFTVideo Stitching using Improved RANSAC and SIFT
Video Stitching using Improved RANSAC and SIFT
 
All projects
All projectsAll projects
All projects
 
ICRA Nathan Piasco
ICRA Nathan PiascoICRA Nathan Piasco
ICRA Nathan Piasco
 
Convolutional Neural Network for pixel-wise skyline detection
Convolutional Neural Network for pixel-wise skyline detectionConvolutional Neural Network for pixel-wise skyline detection
Convolutional Neural Network for pixel-wise skyline detection
 
NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
 
Real time traffic management - challenges and solutions
Real time traffic management - challenges and solutionsReal time traffic management - challenges and solutions
Real time traffic management - challenges and solutions
 
Computer Vision: Feature matching with RANSAC Algorithm
Computer Vision: Feature matching with RANSAC AlgorithmComputer Vision: Feature matching with RANSAC Algorithm
Computer Vision: Feature matching with RANSAC Algorithm
 
crowd counting.pptx
crowd counting.pptxcrowd counting.pptx
crowd counting.pptx
 
Landmark Retrieval & Recognition
Landmark Retrieval & RecognitionLandmark Retrieval & Recognition
Landmark Retrieval & Recognition
 
Deep image retrieval learning global representations for image search
Deep image retrieval  learning global representations for image searchDeep image retrieval  learning global representations for image search
Deep image retrieval learning global representations for image search
 
Towards better performance: phase congruency based face recognition
Towards better performance: phase congruency based face recognitionTowards better performance: phase congruency based face recognition
Towards better performance: phase congruency based face recognition
 
20210226 esa-science-coffee-v2.0
20210226 esa-science-coffee-v2.020210226 esa-science-coffee-v2.0
20210226 esa-science-coffee-v2.0
 

Recently uploaded

Recycled Concrete Aggregate in Construction Part III
Recycled Concrete Aggregate in Construction Part IIIRecycled Concrete Aggregate in Construction Part III
Recycled Concrete Aggregate in Construction Part III
Aditya Rajan Patra
 
2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf
2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf
2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf
Yasser Mahgoub
 
Engineering Drawings Lecture Detail Drawings 2014.pdf
Engineering Drawings Lecture Detail Drawings 2014.pdfEngineering Drawings Lecture Detail Drawings 2014.pdf
Engineering Drawings Lecture Detail Drawings 2014.pdf
abbyasa1014
 
5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...
5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...
5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...
ihlasbinance2003
 
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
insn4465
 
Casting-Defect-inSlab continuous casting.pdf
Casting-Defect-inSlab continuous casting.pdfCasting-Defect-inSlab continuous casting.pdf
Casting-Defect-inSlab continuous casting.pdf
zubairahmad848137
 
Electric vehicle and photovoltaic advanced roles in enhancing the financial p...
Electric vehicle and photovoltaic advanced roles in enhancing the financial p...Electric vehicle and photovoltaic advanced roles in enhancing the financial p...
Electric vehicle and photovoltaic advanced roles in enhancing the financial p...
IJECEIAES
 
Literature Review Basics and Understanding Reference Management.pptx
Literature Review Basics and Understanding Reference Management.pptxLiterature Review Basics and Understanding Reference Management.pptx
Literature Review Basics and Understanding Reference Management.pptx
Dr Ramhari Poudyal
 
International Conference on NLP, Artificial Intelligence, Machine Learning an...
International Conference on NLP, Artificial Intelligence, Machine Learning an...International Conference on NLP, Artificial Intelligence, Machine Learning an...
International Conference on NLP, Artificial Intelligence, Machine Learning an...
gerogepatton
 
Manufacturing Process of molasses based distillery ppt.pptx
Manufacturing Process of molasses based distillery ppt.pptxManufacturing Process of molasses based distillery ppt.pptx
Manufacturing Process of molasses based distillery ppt.pptx
Madan Karki
 
The Python for beginners. This is an advance computer language.
The Python for beginners. This is an advance computer language.The Python for beginners. This is an advance computer language.
The Python for beginners. This is an advance computer language.
sachin chaurasia
 
Advanced control scheme of doubly fed induction generator for wind turbine us...
Advanced control scheme of doubly fed induction generator for wind turbine us...Advanced control scheme of doubly fed induction generator for wind turbine us...
Advanced control scheme of doubly fed induction generator for wind turbine us...
IJECEIAES
 
BPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdf
BPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdfBPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdf
BPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdf
MIGUELANGEL966976
 
Iron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdf
Iron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdfIron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdf
Iron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdf
RadiNasr
 
学校原版美国波士顿大学毕业证学历学位证书原版一模一样
学校原版美国波士顿大学毕业证学历学位证书原版一模一样学校原版美国波士顿大学毕业证学历学位证书原版一模一样
学校原版美国波士顿大学毕业证学历学位证书原版一模一样
171ticu
 
Harnessing WebAssembly for Real-time Stateless Streaming Pipelines
Harnessing WebAssembly for Real-time Stateless Streaming PipelinesHarnessing WebAssembly for Real-time Stateless Streaming Pipelines
Harnessing WebAssembly for Real-time Stateless Streaming Pipelines
Christina Lin
 
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMSA SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
IJNSA Journal
 
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Sinan KOZAK
 
New techniques for characterising damage in rock slopes.pdf
New techniques for characterising damage in rock slopes.pdfNew techniques for characterising damage in rock slopes.pdf
New techniques for characterising damage in rock slopes.pdf
wisnuprabawa3
 
basic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdfbasic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdf
NidhalKahouli2
 

Recently uploaded (20)

Recycled Concrete Aggregate in Construction Part III
Recycled Concrete Aggregate in Construction Part IIIRecycled Concrete Aggregate in Construction Part III
Recycled Concrete Aggregate in Construction Part III
 
2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf
2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf
2008 BUILDING CONSTRUCTION Illustrated - Ching Chapter 02 The Building.pdf
 
Engineering Drawings Lecture Detail Drawings 2014.pdf
Engineering Drawings Lecture Detail Drawings 2014.pdfEngineering Drawings Lecture Detail Drawings 2014.pdf
Engineering Drawings Lecture Detail Drawings 2014.pdf
 
5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...
5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...
5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...
 
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
 
Casting-Defect-inSlab continuous casting.pdf
Casting-Defect-inSlab continuous casting.pdfCasting-Defect-inSlab continuous casting.pdf
Casting-Defect-inSlab continuous casting.pdf
 
Electric vehicle and photovoltaic advanced roles in enhancing the financial p...
Electric vehicle and photovoltaic advanced roles in enhancing the financial p...Electric vehicle and photovoltaic advanced roles in enhancing the financial p...
Electric vehicle and photovoltaic advanced roles in enhancing the financial p...
 
Literature Review Basics and Understanding Reference Management.pptx
Literature Review Basics and Understanding Reference Management.pptxLiterature Review Basics and Understanding Reference Management.pptx
Literature Review Basics and Understanding Reference Management.pptx
 
International Conference on NLP, Artificial Intelligence, Machine Learning an...
International Conference on NLP, Artificial Intelligence, Machine Learning an...International Conference on NLP, Artificial Intelligence, Machine Learning an...
International Conference on NLP, Artificial Intelligence, Machine Learning an...
 
Manufacturing Process of molasses based distillery ppt.pptx
Manufacturing Process of molasses based distillery ppt.pptxManufacturing Process of molasses based distillery ppt.pptx
Manufacturing Process of molasses based distillery ppt.pptx
 
The Python for beginners. This is an advance computer language.
The Python for beginners. This is an advance computer language.The Python for beginners. This is an advance computer language.
The Python for beginners. This is an advance computer language.
 
Advanced control scheme of doubly fed induction generator for wind turbine us...
Advanced control scheme of doubly fed induction generator for wind turbine us...Advanced control scheme of doubly fed induction generator for wind turbine us...
Advanced control scheme of doubly fed induction generator for wind turbine us...
 
BPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdf
BPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdfBPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdf
BPV-GUI-01-Guide-for-ASME-Review-Teams-(General)-10-10-2023.pdf
 
Iron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdf
Iron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdfIron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdf
Iron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdf
 
学校原版美国波士顿大学毕业证学历学位证书原版一模一样
学校原版美国波士顿大学毕业证学历学位证书原版一模一样学校原版美国波士顿大学毕业证学历学位证书原版一模一样
学校原版美国波士顿大学毕业证学历学位证书原版一模一样
 
Harnessing WebAssembly for Real-time Stateless Streaming Pipelines
Harnessing WebAssembly for Real-time Stateless Streaming PipelinesHarnessing WebAssembly for Real-time Stateless Streaming Pipelines
Harnessing WebAssembly for Real-time Stateless Streaming Pipelines
 
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMSA SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
 
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
 
New techniques for characterising damage in rock slopes.pdf
New techniques for characterising damage in rock slopes.pdfNew techniques for characterising damage in rock slopes.pdf
New techniques for characterising damage in rock slopes.pdf
 
basic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdfbasic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdf
 

An accurate retrieval through R-MAC+ descriptors for landmark recognition

  • 1. An accurate retrieval through R-MAC+ descriptors for landmark recognition Federico Magliani, Andrea Prati ICDSC 2018 – Eindhoven, Netherlands – 3-4 September 2018
  • 2. Agenda 2 ➢ Motivations ➢ Summary of contributions ➢ Related works ➢ Introduction to R-MAC descriptors ➢ Proposed approach (R-MAC+) ➢ Experimental results ➢ Conclusions
  • 3. Motivations 3 Landmark Recognition problem ➢ Try to understand what’s is in front of you and retrieve similar images. ➢ Semantic gap: for a human, this task is pretty simple thanks to personal experience, but a computer can use only the info available in the images. ➢ It is far from being solved (viewpoint, illumination conditions, image resolution, ...).
  • 4. Motivations 4 ➢ Challenges ○ High accuracy retrieval (precision) ○ Fast research (response to query) ○ Reduced memory occupied (mobile friendly) ○ Work well with big data (>1M data) ➢ Possible applications ○ Augmented reality (tourism) ○ Person Re-ID (video-surveillance) ○ Online clothes search (fashion)
  • 5. Agenda 5 ➢ Motivations ➢ Summary of contributions ➢ Related works ➢ Introduction to R-MAC descriptors ➢ Proposed approach (R-MAC+) ➢ Experimental results ➢ Conclusions and Future Works
  • 6. Summary of contributions 6 ➢ a new region detector for CNN feature maps implemented through grids, that respect the aspect ratio of the images. ➢ an improvement on the effectiveness of the multi-resolution approach for R-MAC descriptors. ➢ a novel retrieval method for checking the similarities between query descriptors and regions of database R-MAC descriptors. It allows to outperform the results of R-MAC descriptors on Oxford5k and Paris6k by +7% and +3%.
  • 7. Agenda 7 ➢ Motivations ➢ Summary of contribution ➢ Related works ➢ Introduction to R-MAC descriptors ➢ Proposed approach (R-MAC+) ➢ Experimental results ➢ Conclusions
  • 8. Related works 8 ➢ Bag of Words (BoW): first method for solving the problem (different techniques: vocabulary tree, …). ➢ VLAD: similar to BoW, but using the residual of the descriptors (=feature descriptor - closest centers in the vocabulary). ➢ CNN based: extract features from intermediate layers of CNN architectures and then apply previous embedding techniques (BLCF, ...). ➢ MAC: max pooling applied on CNN features ➢ R-MAC: regional MAC descriptors created through the application of a rigid-grid mechanism
  • 9. Agenda 9 ➢ Motivations ➢ Summary of contributions ➢ Related works ➢ Introduction to R-MAC descriptors ➢ Proposed approach (R-MAC+) ➢ Experimental results ➢ Conclusions
  • 10. R-MAC (Regional MAC) descriptors 10 Considering a rectangular region R ⊆ Ω = (1,W) x (1,H), and define the regional feature vector: fR = (fR,1 ...fR,i ...fR,K )T where fR,i = max Xi (p) is the maximum activation of the ith channel on the considered region. Then we calculate the feature vector associated with each region, and post-process it with l2 -normalization, PCA-whitening and l2 -normalization. We combine the collection of regional feature vectors into a single image vector by summing them and l2 -normalizing in the end. We define the response maps and sample square regions at L different scales ➢ at the largest scale (l=1), the region size is determined to be as large as possible (height = width = min(W,H)) ➢ at every other scale l, we uniformly sample l x (l+m-1) regions of width 2min(W,H)/(l+1). (with m=2)
  • 11. R-MAC (Regional MAC) descriptors 11 Settings: ➢ Fully convolutional off-the-shelf VGG16 ➢ Pool5 ➢ Spatial Max pooling ➢ High Resolution images ➢ Global descriptor based on aggregating region vectors ➢ Sliding window approach Tolias et al. Particular object retrieval with integral max-pooling of CNN activations. arXiv 2015.
  • 12. Agenda 12 ➢ Motivations ➢ Summary of contributions ➢ Related works ➢ Introduction to R-MAC descriptors ➢ Proposed approach (R-MAC+) ➢ Experimental results ➢ Conclusions
  • 13. Proposed approach: R-MAC+ New multi-resolution approach: the images are resized of +25%,-25%, 0% on the largest size, respecting the aspect ratio of the image. ➢ This strategy is an alternative of the first multi-resolution approach, that resized the image to a fixed size: 550px, 800px and 1050 on the largest size, retaining the aspect ratio of the image. ➢ This strategy should allow to augment the dimensions of the feature maps in order to have more features and therefore local maxima than the previous multi-resolution R-MAC. This approach is connected to the new region detector, that detects a reduced number of regions (15) instead of the 20 of the original one. 13
  • 14. Proposed approach: R-MAC+ 14 A new mechanism for region detection in the CNN feature maps (15 regions) ● l=0 → 1 region covering entirely the image; ● l=1 → 2 square regions (widthRegion = heightRegion = min(H,W)); ● l=2 → 6 rect regions (widthRegion = heightRegion =⌈2*min(W,H)/(l+1))⌉, arranged along the horizontal axis (width and height of the regions are adapted to cover all the image); ● l=3 → 6 rect regions (widthRegion = heightRegion= ⌈2*min(W,H)/(l+2))⌉, arranged along the vertical axis (width and height of the regions are adapted to cover all the image).
  • 15. Proposed approach: R-MAC+ 15 A new retrieval method based on db regions (MAC descriptors of the database images) and the R-MAC descriptors of the query images (+7% on Oxford5k and +4% on Paris6k than previous results)
  • 16. Agenda 16 ➢ Motivations ➢ Summary of contributions ➢ Related works ➢ Introduction to R-MAC descriptors ➢ Proposed approach (R-MAC+) ➢ Experimental results ➢ Conclusions
  • 17. Datasets and evaluation metric Datasets: ➢ Holidays (1491 images: 500 classes, 500 queries). ➢ Oxford5k (5063 images, 11 classes, 55 queries). ➢ Paris6k (6412 images, 11 classes, 55 queries). Evaluation metric: ➢ mAP (mean Average Precision) → mean of Average Precision scores (correct results) for each query, based on the position in the ranking. 17
  • 18. Results 18 Method Network Holidays (original/rotated) Oxf5k Paris6k MAC VGG19 76.26 % 57.44 % 73.15 % R-MAC VGG19 87.65 % 65.56 % 82.80 % R-MAC ResNet50 92.55 % 71.77 % 83.31 % M-R R-MAC+ ResNet50 94.63 % / 95.58 % 78.88 % 88.63 % M-R R-MAC+ with retrieval based on db regions ResNet50 94.37 % / 95.87 % 85.39 % 91.90 %
  • 19. Results after QE application 19 Method Network Holidays (original/rotated) Oxf5k Paris6k M-R R-MAC+ ResNet50 94.97 % / 95.97 % 86.45 % 92.01 % M-R R-MAC+ with retrieval based on db regions ResNet50 94.42 % / 96.05 % 87.92 % 93.64 % M-R R-MAC+ with retrieval based on db regions and query expansion based on db regions ResNet50 94.28 % / 95.91 % 88.78 % 92.30 %
  • 20. Comparison with the state of the art 20
  • 21. Agenda 21 ➢ Motivations ➢ Summary of contributions ➢ Related works ➢ Introduction to R-MAC descriptors ➢ Proposed approach (R-MAC+) ➢ Experimental results ➢ Conclusions
  • 22. Conclusions ➢ We propose different improvements on R-MAC descriptors in order to make the retrieval very accurate. ○ A multi-resolution approach, that uses bigger feature maps than the previous one. ○ A new region detector with the use of adaptable grids allows to catch more local maxima. ○ A novel retrieval method based on db regions that highly boosts the performance on Oxford5k and Paris6k. ➢ The proposed method outperforms the state of the art on Holidays, both on the original and rotated version. Also it outperforms the state-of-the-art results on some other public benchmarks without the fine-tuning application. 22
  • 23. Thank you for your attention! questions? http://implab.ce.unipr.it 23