SlideShare ist ein Scribd-Unternehmen logo
1 von 29
1. S. Bai et al., Scalable Person Re-identification on Supervised Smoothed Manifold.
CVPR2017 spotlight
2. Y. Sun et al., Deep Representation Learning: a Similarity Smoothing Perspective
3. Y. Shen et al., Person Re-identification with Deep Similarity-Guided Graph Neural
Network. ECCV 2018 accepted
Keywords: Smooth Similarity, Smoothed Manifold, Graph Neural Network
孙奕帆
ReID中的相似性平滑约束
Scalable Re-ID on Supervised Smoothed Manifold.
Reference:
1. D. Zhou et al., Learning with local and global consistency, NIPS2003
2. S. Bai et al., Scalable Person Re-identification on Supervised Smoothed Manifold. CVPR2017
何为相似性不平滑
PA: A1 A2
B2
假设某特征空间中存在以下关系:
A1 close to B1
A2 close to B2
现考察样本对内部的样本距离,假设:
A1 close to A2
B1 far from B2
PB : B1
PA close to PB
互斥于 PA close to PB
一个极端的例子:假设2组样本对PA=(A1,A2)及PB=(B1,B2)
如何施加相似性平滑约束
k
i
l
j
Qki: 样本k,i间的相似性 Qlj: 样本l,j间的相似性
P(ki→lj) 样本对ki到样本对lj的状态转移概率
Lki: 样本k,i是否属于同一ID
正样对为1,负样对为0,L视为硬化的相似性,故可与Q进行加法运算
直观意义:k,i之间的相似性Qk i“吸收”所有其它样本的相似性
Ql j, l,j∈{1,2,…,N} ,
吸收强度由样本对{l,j}、{k,i}之间的状态转移概率P(ki→lj)决定
Scalable Re-ID on Supervised Smoothed Manifold.
如何施加相似性平滑约束
k
i
l
j
Qki: 样本k,i间的相似性 Qlj: 样本l,j间的相似性
P(ki→lj) 样本对ki到样本对lj的状态转移概率
Lki: 样本k,i是否属于同一ID
正样对为1,负样对为0,L视为硬化的相似性,故可与Q进行加法运算
直观意义:k,i之间的相似性Qk i“吸收”所有其它样本的相似性
Ql j, l,j∈{1,2,…,N} ,
吸收强度由样本对{l,j}、{k,i}之间的状态转移概率P(ki→lj)决定
Scalable Re-ID on Supervised Smoothed Manifold.
真的是
所有吗?
相似性吸收强度
2
W expij
dij

  
      
一个常用定义:
控制半径
越小,越关注局部:欧氏距离近的样本对,其W绝对占优,导致吸收仅在较小局部发生
越大,越关注全局:欧氏距离较远的样本对,其强度仍然能被吸收
Scalable Re-ID on Supervised Smoothed Manifold.
SSM的一些问题:
采用后处理方式,增加测试阶段时间。在整个样本空间
(train+test)进行相似性传递,复杂度接近O(N4) (N为样本数),
难以在大训练集上学习。且利用了gallery信息
Deep Representation Learning:
a Similarity Smoothing Perspective
Yifan Sun, Liang Zheng, Qin Xu, Zhongdao Wang, Shengjin Wang
Motivation----smooth similarity
• Has been valued in semi-supervised learning or transductive inference
• Has not been explored in deep representation learning under fully
supervision
A1
A2
B1
B2
Pair I Pair II
similar dissimilar
A1
A2
B1
B2
Pair I Pair II
A1
A2
B1
B2
Pair I Pair II
(a) smooth (b) smooth (c) unsmooth
Our Contribution
• We introduce the smooth similarity constraint, which is traditionally utilized in
semi-supervised learning, to deep representation learning under the fully
supervised manner
• We define an evaluation protocol to measure similarity smoothness and
transform it to a Similarity Smoothing Regularizer (SSR).
• We demonstrate though extensive experiments on four fine-grained retrieval
datasets, that similarity smoothing is beneficial towards more discriminative
representation.
Proposed Method
• A revisit to Smooth Similarity Constraint
• Similarity Smoothness Indicator
• Similarity Smoothing Regularizer (SSR)
• The similarity measure W
• A light edition of SSR for efficient training
Proposed Method
• A revisit to Smooth Similarity Constraint
Affinity value which is initialized with W
Proposed Method
• A revisit to Smooth Similarity Constraint
Affinity value which is initialized with W
Wij: the similarity value calculated with similarity measure
which may be heuristically defined
Proposed Method
• Similarity Smoothness Indicator
Proposed Method
• Similarity Smoothness Indicator
A weighted mean of the similarity variations
between sample pairs
Proposed Method
• Similarity Smoothing Regularizer (SSR)
• Takes the same formula as the Similarity Smoothness Indicator
• To be evaluated within the training mini-batch
• Essentially enforces the similarity not to change too much between nearby pairs
• May achieve a optimum Solution A, compromising the discriminative ability
• So, it is important to combine a metric loss
Proposed Method
• The similarity measure W
• Cosine similarity
• Gaussian similarity
RBF kernel width
Proposed Method
• The similarity measure W
• Gaussian similarity
RBF kernel width impacts on the optimization of SSR
Inappropriate settings will lead SSR to approximate
another optimum Solution B, decreasing the retrieval accuracy:
Proposed Method
• A light edition of SSR for efficient training
When adopting the N-pairs sampling strategy, the computational cost is reduced by V^4
(1/4096 when 8 instances for a same identity)
while bringing little impact on the retrieval accuracy
LSSR focuses on inter-class similarity smoothness
Task----Fine grained Retrieval
• Person Re-identification
• Birds & Cars Retrieval (王重道)
CUB-200 CARS196
Experiments-reID
Experiments-birds and cars retrieval
Baseline 4: contrastive loss alone
Experiments-Mechanism Study
Impact of Similarity Measure W
1)under cosine similarity
2) The effectiveness of using Gaussian Simlarity
Depends
3) A interesting observation:
Only when SSR construct a competing effect
with the cooperating metric loss, (Solution A),
the accuracy increases
Experiments-Mechanism Study
Visualization
SGGNN----propagating features within mini-batch
Element-wise subtraction
and square operation
FC+sigmoid (positive pair1
negative pair0)
Absorbing difference feature
“d” from other pairs.
The absorbing weight W is
determined by the transition
probability
SGGNN----propagating features within mini-batch
The message (or feature) to propagate
is a learnable t = 2FC+BN on “d”, rather
than “d” itself
SGGNN----propagating features within mini-batch
• Performance
• The similarity scores
are predicted by SGGNN
---time consuming
SGGNN----propagating features within mini-batch
Connection & Difference between SSR and SGGNN
• Both method employ smooth similarity constraint on the training dataset (instead
of on the training + testing)
A1
A2
B1
B2
Pair I Pair II
SGGNN: propagating
features within
triplet (special case
of quadruple)
SSR:
propagating
similarities
between any
sample pairs
(quadruple)
A1
A2
B2
Pair I Pair II

Weitere ähnliche Inhalte

Ähnlich wie Smoothed manifold

RBHF_SDM_2011_Jie
RBHF_SDM_2011_JieRBHF_SDM_2011_Jie
RBHF_SDM_2011_JieMDO_Lab
 
Recommender system
Recommender systemRecommender system
Recommender systemBhumi Patel
 
A comparison of SIFT, PCA-SIFT and SURF
A comparison of SIFT, PCA-SIFT and SURFA comparison of SIFT, PCA-SIFT and SURF
A comparison of SIFT, PCA-SIFT and SURFCSCJournals
 
Zero-shot Image Recognition Using Relational Matching, Adaptation and Calibra...
Zero-shot Image Recognition Using Relational Matching, Adaptation and Calibra...Zero-shot Image Recognition Using Relational Matching, Adaptation and Calibra...
Zero-shot Image Recognition Using Relational Matching, Adaptation and Calibra...Debasmit Das
 
M phil-computer-science-remote-sensing-projects
M phil-computer-science-remote-sensing-projectsM phil-computer-science-remote-sensing-projects
M phil-computer-science-remote-sensing-projectsVijay Karan
 
M phil-computer-science-remote-sensing-projects
M phil-computer-science-remote-sensing-projectsM phil-computer-science-remote-sensing-projects
M phil-computer-science-remote-sensing-projectsVijay Karan
 
M.Phil Computer Science Remote Sensing Projects
M.Phil Computer Science Remote Sensing ProjectsM.Phil Computer Science Remote Sensing Projects
M.Phil Computer Science Remote Sensing ProjectsVijay Karan
 
M.E Computer Science Remote Sensing Projects
M.E Computer Science Remote Sensing ProjectsM.E Computer Science Remote Sensing Projects
M.E Computer Science Remote Sensing ProjectsVijay Karan
 
Probabilistic Collaborative Filtering with Negative Cross Entropy
Probabilistic Collaborative Filtering with Negative Cross EntropyProbabilistic Collaborative Filtering with Negative Cross Entropy
Probabilistic Collaborative Filtering with Negative Cross EntropyAlejandro Bellogin
 
Presentation File of paper "Leveraging Normalization Layer in Adapters With P...
Presentation File of paper "Leveraging Normalization Layer in Adapters With P...Presentation File of paper "Leveraging Normalization Layer in Adapters With P...
Presentation File of paper "Leveraging Normalization Layer in Adapters With P...dyyjkd
 
Stopped Training and Other Remedies for OverFITtting
Stopped Training and Other Remedies for OverFITttingStopped Training and Other Remedies for OverFITtting
Stopped Training and Other Remedies for OverFITttingESCOM
 
Supervised WSD Using Master- Slave Voting Technique
Supervised WSD Using Master- Slave Voting TechniqueSupervised WSD Using Master- Slave Voting Technique
Supervised WSD Using Master- Slave Voting Techniqueiosrjce
 
Nonlinear Exponential Regularization : An Improved Version of Regularization ...
Nonlinear Exponential Regularization : An Improved Version of Regularization ...Nonlinear Exponential Regularization : An Improved Version of Regularization ...
Nonlinear Exponential Regularization : An Improved Version of Regularization ...Seoung-Ho Choi
 
November, 2006 CCKM'06 1
November, 2006 CCKM'06 1 November, 2006 CCKM'06 1
November, 2006 CCKM'06 1 butest
 
An Analytical Comparison of Different Regularization Parameter Selection Meth...
An Analytical Comparison of Different Regularization Parameter Selection Meth...An Analytical Comparison of Different Regularization Parameter Selection Meth...
An Analytical Comparison of Different Regularization Parameter Selection Meth...SARADINDU SENGUPTA
 
Computing Neighbourhoods with Language Models in a Collaborative Filtering Sc...
Computing Neighbourhoods with Language Models in a Collaborative Filtering Sc...Computing Neighbourhoods with Language Models in a Collaborative Filtering Sc...
Computing Neighbourhoods with Language Models in a Collaborative Filtering Sc...Daniel Valcarce
 
Learning On The Border:Active Learning in Imbalanced classification Data
Learning On The Border:Active Learning in Imbalanced classification DataLearning On The Border:Active Learning in Imbalanced classification Data
Learning On The Border:Active Learning in Imbalanced classification Data萍華 楊
 
GAN(と強化学習との関係)
GAN(と強化学習との関係)GAN(と強化学習との関係)
GAN(と強化学習との関係)Masahiro Suzuki
 

Ähnlich wie Smoothed manifold (20)

RBHF_SDM_2011_Jie
RBHF_SDM_2011_JieRBHF_SDM_2011_Jie
RBHF_SDM_2011_Jie
 
Recommender system
Recommender systemRecommender system
Recommender system
 
A comparison of SIFT, PCA-SIFT and SURF
A comparison of SIFT, PCA-SIFT and SURFA comparison of SIFT, PCA-SIFT and SURF
A comparison of SIFT, PCA-SIFT and SURF
 
Zero-shot Image Recognition Using Relational Matching, Adaptation and Calibra...
Zero-shot Image Recognition Using Relational Matching, Adaptation and Calibra...Zero-shot Image Recognition Using Relational Matching, Adaptation and Calibra...
Zero-shot Image Recognition Using Relational Matching, Adaptation and Calibra...
 
M phil-computer-science-remote-sensing-projects
M phil-computer-science-remote-sensing-projectsM phil-computer-science-remote-sensing-projects
M phil-computer-science-remote-sensing-projects
 
M phil-computer-science-remote-sensing-projects
M phil-computer-science-remote-sensing-projectsM phil-computer-science-remote-sensing-projects
M phil-computer-science-remote-sensing-projects
 
M.Phil Computer Science Remote Sensing Projects
M.Phil Computer Science Remote Sensing ProjectsM.Phil Computer Science Remote Sensing Projects
M.Phil Computer Science Remote Sensing Projects
 
M.E Computer Science Remote Sensing Projects
M.E Computer Science Remote Sensing ProjectsM.E Computer Science Remote Sensing Projects
M.E Computer Science Remote Sensing Projects
 
Probabilistic Collaborative Filtering with Negative Cross Entropy
Probabilistic Collaborative Filtering with Negative Cross EntropyProbabilistic Collaborative Filtering with Negative Cross Entropy
Probabilistic Collaborative Filtering with Negative Cross Entropy
 
Presentation File of paper "Leveraging Normalization Layer in Adapters With P...
Presentation File of paper "Leveraging Normalization Layer in Adapters With P...Presentation File of paper "Leveraging Normalization Layer in Adapters With P...
Presentation File of paper "Leveraging Normalization Layer in Adapters With P...
 
Stopped Training and Other Remedies for OverFITtting
Stopped Training and Other Remedies for OverFITttingStopped Training and Other Remedies for OverFITtting
Stopped Training and Other Remedies for OverFITtting
 
J017256674
J017256674J017256674
J017256674
 
Supervised WSD Using Master- Slave Voting Technique
Supervised WSD Using Master- Slave Voting TechniqueSupervised WSD Using Master- Slave Voting Technique
Supervised WSD Using Master- Slave Voting Technique
 
panel regression.pptx
panel regression.pptxpanel regression.pptx
panel regression.pptx
 
Nonlinear Exponential Regularization : An Improved Version of Regularization ...
Nonlinear Exponential Regularization : An Improved Version of Regularization ...Nonlinear Exponential Regularization : An Improved Version of Regularization ...
Nonlinear Exponential Regularization : An Improved Version of Regularization ...
 
November, 2006 CCKM'06 1
November, 2006 CCKM'06 1 November, 2006 CCKM'06 1
November, 2006 CCKM'06 1
 
An Analytical Comparison of Different Regularization Parameter Selection Meth...
An Analytical Comparison of Different Regularization Parameter Selection Meth...An Analytical Comparison of Different Regularization Parameter Selection Meth...
An Analytical Comparison of Different Regularization Parameter Selection Meth...
 
Computing Neighbourhoods with Language Models in a Collaborative Filtering Sc...
Computing Neighbourhoods with Language Models in a Collaborative Filtering Sc...Computing Neighbourhoods with Language Models in a Collaborative Filtering Sc...
Computing Neighbourhoods with Language Models in a Collaborative Filtering Sc...
 
Learning On The Border:Active Learning in Imbalanced classification Data
Learning On The Border:Active Learning in Imbalanced classification DataLearning On The Border:Active Learning in Imbalanced classification Data
Learning On The Border:Active Learning in Imbalanced classification Data
 
GAN(と強化学習との関係)
GAN(と強化学習との関係)GAN(と強化学習との関係)
GAN(と強化学習との関係)
 

Mehr von 哲东 郑

Deep learning for person re-identification
Deep learning for person re-identificationDeep learning for person re-identification
Deep learning for person re-identification哲东 郑
 
Cross-domain complementary learning with synthetic data for multi-person part...
Cross-domain complementary learning with synthetic data for multi-person part...Cross-domain complementary learning with synthetic data for multi-person part...
Cross-domain complementary learning with synthetic data for multi-person part...哲东 郑
 
Visual saliency
Visual saliencyVisual saliency
Visual saliency哲东 郑
 
Image Synthesis From Reconfigurable Layout and Style
Image Synthesis From Reconfigurable Layout and StyleImage Synthesis From Reconfigurable Layout and Style
Image Synthesis From Reconfigurable Layout and Style哲东 郑
 
Polysemous Visual-Semantic Embedding for Cross-Modal Retrieval
Polysemous Visual-Semantic Embedding for Cross-Modal RetrievalPolysemous Visual-Semantic Embedding for Cross-Modal Retrieval
Polysemous Visual-Semantic Embedding for Cross-Modal Retrieval哲东 郑
 
Weijian image retrieval
Weijian image retrievalWeijian image retrieval
Weijian image retrieval哲东 郑
 
Scops self supervised co-part segmentation
Scops self supervised co-part segmentationScops self supervised co-part segmentation
Scops self supervised co-part segmentation哲东 郑
 
Video object detection
Video object detectionVideo object detection
Video object detection哲东 郑
 
C2 ae open set recognition
C2 ae open set recognitionC2 ae open set recognition
C2 ae open set recognition哲东 郑
 
Sota semantic segmentation
Sota semantic segmentationSota semantic segmentation
Sota semantic segmentation哲东 郑
 
Deep randomized embedding
Deep randomized embeddingDeep randomized embedding
Deep randomized embedding哲东 郑
 
Semantic Image Synthesis with Spatially-Adaptive Normalization
Semantic Image Synthesis with Spatially-Adaptive NormalizationSemantic Image Synthesis with Spatially-Adaptive Normalization
Semantic Image Synthesis with Spatially-Adaptive Normalization哲东 郑
 
Instance level facial attributes transfer with geometry-aware flow
Instance level facial attributes transfer with geometry-aware flowInstance level facial attributes transfer with geometry-aware flow
Instance level facial attributes transfer with geometry-aware flow哲东 郑
 
Learning to adapt structured output space for semantic
Learning to adapt structured output space for semanticLearning to adapt structured output space for semantic
Learning to adapt structured output space for semantic哲东 郑
 
Unsupervised Learning of Object Landmarks through Conditional Image Generation
Unsupervised Learning of Object Landmarks through Conditional Image GenerationUnsupervised Learning of Object Landmarks through Conditional Image Generation
Unsupervised Learning of Object Landmarks through Conditional Image Generation哲东 郑
 
Graph based global reasoning networks
Graph based global reasoning networks Graph based global reasoning networks
Graph based global reasoning networks 哲东 郑
 

Mehr von 哲东 郑 (20)

Deep learning for person re-identification
Deep learning for person re-identificationDeep learning for person re-identification
Deep learning for person re-identification
 
Cross-domain complementary learning with synthetic data for multi-person part...
Cross-domain complementary learning with synthetic data for multi-person part...Cross-domain complementary learning with synthetic data for multi-person part...
Cross-domain complementary learning with synthetic data for multi-person part...
 
Step zhedong
Step zhedongStep zhedong
Step zhedong
 
Visual saliency
Visual saliencyVisual saliency
Visual saliency
 
Image Synthesis From Reconfigurable Layout and Style
Image Synthesis From Reconfigurable Layout and StyleImage Synthesis From Reconfigurable Layout and Style
Image Synthesis From Reconfigurable Layout and Style
 
Polysemous Visual-Semantic Embedding for Cross-Modal Retrieval
Polysemous Visual-Semantic Embedding for Cross-Modal RetrievalPolysemous Visual-Semantic Embedding for Cross-Modal Retrieval
Polysemous Visual-Semantic Embedding for Cross-Modal Retrieval
 
Weijian image retrieval
Weijian image retrievalWeijian image retrieval
Weijian image retrieval
 
Scops self supervised co-part segmentation
Scops self supervised co-part segmentationScops self supervised co-part segmentation
Scops self supervised co-part segmentation
 
Video object detection
Video object detectionVideo object detection
Video object detection
 
Center nets
Center netsCenter nets
Center nets
 
C2 ae open set recognition
C2 ae open set recognitionC2 ae open set recognition
C2 ae open set recognition
 
Sota semantic segmentation
Sota semantic segmentationSota semantic segmentation
Sota semantic segmentation
 
Deep randomized embedding
Deep randomized embeddingDeep randomized embedding
Deep randomized embedding
 
Semantic Image Synthesis with Spatially-Adaptive Normalization
Semantic Image Synthesis with Spatially-Adaptive NormalizationSemantic Image Synthesis with Spatially-Adaptive Normalization
Semantic Image Synthesis with Spatially-Adaptive Normalization
 
Instance level facial attributes transfer with geometry-aware flow
Instance level facial attributes transfer with geometry-aware flowInstance level facial attributes transfer with geometry-aware flow
Instance level facial attributes transfer with geometry-aware flow
 
Learning to adapt structured output space for semantic
Learning to adapt structured output space for semanticLearning to adapt structured output space for semantic
Learning to adapt structured output space for semantic
 
Unsupervised Learning of Object Landmarks through Conditional Image Generation
Unsupervised Learning of Object Landmarks through Conditional Image GenerationUnsupervised Learning of Object Landmarks through Conditional Image Generation
Unsupervised Learning of Object Landmarks through Conditional Image Generation
 
Graph based global reasoning networks
Graph based global reasoning networks Graph based global reasoning networks
Graph based global reasoning networks
 
Style gan
Style ganStyle gan
Style gan
 
Vi2vi
Vi2viVi2vi
Vi2vi
 

Kürzlich hochgeladen

Unleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubUnleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubKalema Edgar
 
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostZilliz
 
What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024Stephanie Beckett
 
My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024The Digital Insurer
 
Training state-of-the-art general text embedding
Training state-of-the-art general text embeddingTraining state-of-the-art general text embedding
Training state-of-the-art general text embeddingZilliz
 
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticsKotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticscarlostorres15106
 
Story boards and shot lists for my a level piece
Story boards and shot lists for my a level pieceStory boards and shot lists for my a level piece
Story boards and shot lists for my a level piececharlottematthew16
 
Unraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfUnraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfAlex Barbosa Coqueiro
 
DevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsDevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsSergiu Bodiu
 
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationRidwan Fadjar
 
"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr BaganFwdays
 
AI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsAI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsMemoori
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...Fwdays
 
"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii SoldatenkoFwdays
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024BookNet Canada
 
Streamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupStreamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupFlorian Wilhelm
 
Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!Manik S Magar
 
The Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfThe Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfSeasiaInfotech2
 
Gen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdfGen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdfAddepto
 
Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Scott Keck-Warren
 

Kürzlich hochgeladen (20)

Unleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubUnleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding Club
 
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
 
What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024
 
My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024
 
Training state-of-the-art general text embedding
Training state-of-the-art general text embeddingTraining state-of-the-art general text embedding
Training state-of-the-art general text embedding
 
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticsKotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
 
Story boards and shot lists for my a level piece
Story boards and shot lists for my a level pieceStory boards and shot lists for my a level piece
Story boards and shot lists for my a level piece
 
Unraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfUnraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdf
 
DevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsDevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platforms
 
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 Presentation
 
"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan
 
AI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsAI as an Interface for Commercial Buildings
AI as an Interface for Commercial Buildings
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
 
"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
 
Streamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupStreamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project Setup
 
Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!
 
The Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfThe Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdf
 
Gen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdfGen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdf
 
Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024
 

Smoothed manifold

  • 1. 1. S. Bai et al., Scalable Person Re-identification on Supervised Smoothed Manifold. CVPR2017 spotlight 2. Y. Sun et al., Deep Representation Learning: a Similarity Smoothing Perspective 3. Y. Shen et al., Person Re-identification with Deep Similarity-Guided Graph Neural Network. ECCV 2018 accepted Keywords: Smooth Similarity, Smoothed Manifold, Graph Neural Network 孙奕帆 ReID中的相似性平滑约束
  • 2. Scalable Re-ID on Supervised Smoothed Manifold. Reference: 1. D. Zhou et al., Learning with local and global consistency, NIPS2003 2. S. Bai et al., Scalable Person Re-identification on Supervised Smoothed Manifold. CVPR2017 何为相似性不平滑 PA: A1 A2 B2 假设某特征空间中存在以下关系: A1 close to B1 A2 close to B2 现考察样本对内部的样本距离,假设: A1 close to A2 B1 far from B2 PB : B1 PA close to PB 互斥于 PA close to PB 一个极端的例子:假设2组样本对PA=(A1,A2)及PB=(B1,B2)
  • 3. 如何施加相似性平滑约束 k i l j Qki: 样本k,i间的相似性 Qlj: 样本l,j间的相似性 P(ki→lj) 样本对ki到样本对lj的状态转移概率 Lki: 样本k,i是否属于同一ID 正样对为1,负样对为0,L视为硬化的相似性,故可与Q进行加法运算 直观意义:k,i之间的相似性Qk i“吸收”所有其它样本的相似性 Ql j, l,j∈{1,2,…,N} , 吸收强度由样本对{l,j}、{k,i}之间的状态转移概率P(ki→lj)决定 Scalable Re-ID on Supervised Smoothed Manifold.
  • 4. 如何施加相似性平滑约束 k i l j Qki: 样本k,i间的相似性 Qlj: 样本l,j间的相似性 P(ki→lj) 样本对ki到样本对lj的状态转移概率 Lki: 样本k,i是否属于同一ID 正样对为1,负样对为0,L视为硬化的相似性,故可与Q进行加法运算 直观意义:k,i之间的相似性Qk i“吸收”所有其它样本的相似性 Ql j, l,j∈{1,2,…,N} , 吸收强度由样本对{l,j}、{k,i}之间的状态转移概率P(ki→lj)决定 Scalable Re-ID on Supervised Smoothed Manifold. 真的是 所有吗?
  • 5. 相似性吸收强度 2 W expij dij            一个常用定义: 控制半径 越小,越关注局部:欧氏距离近的样本对,其W绝对占优,导致吸收仅在较小局部发生 越大,越关注全局:欧氏距离较远的样本对,其强度仍然能被吸收 Scalable Re-ID on Supervised Smoothed Manifold.
  • 7. Deep Representation Learning: a Similarity Smoothing Perspective Yifan Sun, Liang Zheng, Qin Xu, Zhongdao Wang, Shengjin Wang
  • 8. Motivation----smooth similarity • Has been valued in semi-supervised learning or transductive inference • Has not been explored in deep representation learning under fully supervision A1 A2 B1 B2 Pair I Pair II similar dissimilar A1 A2 B1 B2 Pair I Pair II A1 A2 B1 B2 Pair I Pair II (a) smooth (b) smooth (c) unsmooth
  • 9. Our Contribution • We introduce the smooth similarity constraint, which is traditionally utilized in semi-supervised learning, to deep representation learning under the fully supervised manner • We define an evaluation protocol to measure similarity smoothness and transform it to a Similarity Smoothing Regularizer (SSR). • We demonstrate though extensive experiments on four fine-grained retrieval datasets, that similarity smoothing is beneficial towards more discriminative representation.
  • 10. Proposed Method • A revisit to Smooth Similarity Constraint • Similarity Smoothness Indicator • Similarity Smoothing Regularizer (SSR) • The similarity measure W • A light edition of SSR for efficient training
  • 11. Proposed Method • A revisit to Smooth Similarity Constraint Affinity value which is initialized with W
  • 12. Proposed Method • A revisit to Smooth Similarity Constraint Affinity value which is initialized with W Wij: the similarity value calculated with similarity measure which may be heuristically defined
  • 13. Proposed Method • Similarity Smoothness Indicator
  • 14. Proposed Method • Similarity Smoothness Indicator A weighted mean of the similarity variations between sample pairs
  • 15. Proposed Method • Similarity Smoothing Regularizer (SSR) • Takes the same formula as the Similarity Smoothness Indicator • To be evaluated within the training mini-batch • Essentially enforces the similarity not to change too much between nearby pairs • May achieve a optimum Solution A, compromising the discriminative ability • So, it is important to combine a metric loss
  • 16. Proposed Method • The similarity measure W • Cosine similarity • Gaussian similarity RBF kernel width
  • 17. Proposed Method • The similarity measure W • Gaussian similarity RBF kernel width impacts on the optimization of SSR Inappropriate settings will lead SSR to approximate another optimum Solution B, decreasing the retrieval accuracy:
  • 18. Proposed Method • A light edition of SSR for efficient training When adopting the N-pairs sampling strategy, the computational cost is reduced by V^4 (1/4096 when 8 instances for a same identity) while bringing little impact on the retrieval accuracy LSSR focuses on inter-class similarity smoothness
  • 19. Task----Fine grained Retrieval • Person Re-identification • Birds & Cars Retrieval (王重道) CUB-200 CARS196
  • 21. Experiments-birds and cars retrieval Baseline 4: contrastive loss alone
  • 22. Experiments-Mechanism Study Impact of Similarity Measure W 1)under cosine similarity 2) The effectiveness of using Gaussian Simlarity Depends 3) A interesting observation: Only when SSR construct a competing effect with the cooperating metric loss, (Solution A), the accuracy increases
  • 24.
  • 25. SGGNN----propagating features within mini-batch Element-wise subtraction and square operation FC+sigmoid (positive pair1 negative pair0) Absorbing difference feature “d” from other pairs. The absorbing weight W is determined by the transition probability
  • 26. SGGNN----propagating features within mini-batch The message (or feature) to propagate is a learnable t = 2FC+BN on “d”, rather than “d” itself
  • 27. SGGNN----propagating features within mini-batch • Performance • The similarity scores are predicted by SGGNN ---time consuming
  • 29. Connection & Difference between SSR and SGGNN • Both method employ smooth similarity constraint on the training dataset (instead of on the training + testing) A1 A2 B1 B2 Pair I Pair II SGGNN: propagating features within triplet (special case of quadruple) SSR: propagating similarities between any sample pairs (quadruple) A1 A2 B2 Pair I Pair II