图像检索相关论文

转载自https://handong1587.github.io/deep_learning/2015/10/09/image-retrieval.html

Image Retrieval

 Published:  09 Oct 2015   Category:  deep_learning

Jump to...

1. Papers
2. Video Indexing / Retrieval
3. Learning to Rank
4. Talks / Slides
5. Datasets
6. Projects
7. Blogs
8. Tutorials

Papers

Using Very Deep Autoencoders for Content-Based Image Retrieval

• intro: ESANN 2011. Alex Krizhevsky, and Geoffrey E. Hinton
• paper: https://www.cs.toronto.edu/~hinton/absps/esann-deep-final.pdf
• paper: http://www.cs.toronto.edu/~fritz/absps/esann-deep-final.pdf

Learning High-level Image Representation for Image Retrieval via Multi-Task DNN using Clickthrough Data

• arxiv: http://arxiv.org/abs/1312.4740
• paper: http://legacy.openreview.net/document/90fc8dad-ad02-4ddc-ab06-e7b55706869d#90fc8dad-ad02-4ddc-ab06-e7b55706869d

Neural Codes for Image Retrieval

• project page: http://sites.skoltech.ru/compvision/projects/neuralcodes/
• arxiv: http://arxiv.org/abs/1404.1777
• github: https://github.com/arbabenko/Spoc

Efficient On-the-fly Category Retrieval using ConvNets and GPUs

• arxiv: http://arxiv.org/abs/1407.4764

Hashing by Deep Learning

• intro: IBM T. J. Watson Research Center
• paper: http://www.ee.columbia.edu/~wliu/WeiLiu_DLHash.pdf

Deep Learning of Binary Hash Codes for Fast Image Retrieval

• intro: CVPR Workshop 2015
• intro: MNIST, CIFAR-10, Yahoo-1M
• paper: http://www.iis.sinica.edu.tw/~kevinlin311.tw/cvprw15.pdf
• github: https://github.com/kevinlin311tw/caffe-cvprw15

Learning visual similarity for product design with convolutional neural networks

• intro: SIGGRAPH 2015
• paper: http://www.cs.cornell.edu/~kb/publications/SIG15ProductNet.pdf
• paper: http://dl.acm.org.sci-hub.cc/citation.cfm?doid=2809654.2766959

Deep Semantic Ranking Based Hashing for Multi-Label Image Retrieval

• intro: CVPR 2015
• arxiv: http://arxiv.org/abs/1501.06272

Exploiting Local Features from Deep Networks for Image Retrieval

• intro: CVPR DeepVision Workshop 2015
• arxiv: https://arxiv.org/abs/1504.05133

Supervised Learning of Semantics-Preserving Hashing via Deep Neural Networks for Large-Scale Image Search

• intro: SSDH
• arxiv: http://arxiv.org/abs/1507.00101
• github: https://github.com/kevinlin311tw/Caffe-DeepBinaryCode

Visual Search at Pinterest

• arxiv: http://arxiv.org/abs/1505.07647
• blog: https://engineering.pinterest.com/blog/introducing-new-way-visually-search-pinterest

Cross-domain Image Retrieval with a Dual Attribute-aware Ranking Network

• intro: ICCV 2015
• intro: DARN, cross-entropy loss, triplet loss
• arxiv: http://arxiv.org/abs/1505.07922

Aggregating Deep Convolutional Features for Image Retrieval

• intro: ICCV 2015
• intro: Sum pooing
• arxiv: http://arxiv.org/abs/1510.07493

Feature Learning based Deep Supervised Hashing with Pairwise Labels

• intro: IJCAI 2016
• arxiv: http://arxiv.org/abs/1511.03855
• code: http://cs.nju.edu.cn/lwj/code/DPSH_code.rar

Particular object retrieval with integral max-pooling of CNN activations

• intro: use max-pooling to aggregate the deep descriptors, R-MAC (regional maximum activation of convolutions)
• arxiv: https://arxiv.org/abs/1511.05879

Group Invariant Deep Representations for Image Instance Retrieval

• arxiv: http://arxiv.org/abs/1601.02093

Where to Buy It: Matching Street Clothing Photos in Online Shops

• intro: ICCV 2015
• hmepage: http://www.tamaraberg.com/street2shop/
• paper: http://www.tamaraberg.com/papers/street2shop.pdf
• paper: http://www.cv-foundation.org/openaccess/content_iccv_2015/html/Kiapour_Where_to_Buy_ICCV_2015_paper.html

Natural Language Object Retrieval

• intro: CVPR 2015
• homepage: http://ronghanghu.com/text_obj_retrieval/
• arxiv: http://arxiv.org/abs/1511.04164
• slides: http://ronghanghu.com/slides/cvpr16_text_obj_retrieval_slides.pdf
• github: https://github.com/ronghanghu/natural-language-object-retrieval
• github: https://github.com/andrewliao11/Natural-Language-Object-Retrieval-tensorflow

Deep Image Retrieval: Learning global representations for image search

• intro: ECCV 2016
• project page: http://www.xrce.xerox.com/Research-Development/Computer-Vision/Learning-Visual-Representations/Deep-Image-Retrieval
• arxiv: https://arxiv.org/abs/1604.01325
• slides: http://www.slideshare.net/xavigiro/deep-image-retrieval-learning-global-representations-for-image-search

Bags of Local Convolutional Features for Scalable Instance Search

• intro: ICMR 2016. Best Poster Award at ICMR 2016.
• project page: https://imatge-upc.github.io/retrieval-2016-icmr/
• arxiv: https://arxiv.org/abs/1604.04653
• github: https://github.com/imatge-upc/retrieval-2016-icmr
• slides: http://www.slideshare.net/xavigiro/convolutional-features-for-instance-search

Faster R-CNN Features for Instance Search

• intro: DeepVision Workshop in CVPR 2016
• homepage: http://imatge-upc.github.io/retrieval-2016-deepvision/
• arxiv: http://arxiv.org/abs/1604.08893
• github: https://github.com/imatge-upc/retrieval-2016-deepvision

Where to Focus: Query Adaptive Matching for Instance Retrieval Using Convolutional Feature Maps

• intro: query adaptive matching (QAM), Feature Map Pooling, Overlapped Spatial Pyramid Pooling (OSPP)
• arxiv: https://arxiv.org/abs/1606.06811

Adversarial Training For Sketch Retrieval

• arxiv: http://arxiv.org/abs/1607.02748

Learning Compact Binary Descriptors with Unsupervised Deep Neural Networks

• intro: CVPR 2016. DeepBit
• paper: http://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Lin_Learning_Compact_Binary_CVPR_2016_paper.pdf
• github: https://github.com/kevinlin311tw/cvpr16-deepbit

Fast Training of Triplet-based Deep Binary Embedding Networks

• intro: CVPR 2016
• paper: http://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Zhuang_Fast_Training_of_CVPR_2016_paper.pdf

Deep Relative Distance Learning: Tell the Difference Between Similar Vehicles

• intro: CVPR 2016
• intro: vehicle re-identification, vehicle retrieval. coupled clusters loss
• paper: http://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Liu_Deep_Relative_Distance_CVPR_2016_paper.pdf

DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations

• intro: CVPR 2016. FashionNet
• project page: http://personal.ie.cuhk.edu.hk/~lz013/projects/DeepFashion.html
• paper: http://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Liu_DeepFashion_Powering_Robust_CVPR_2016_paper.pdf

CNN Image Retrieval Learns from BoW: Unsupervised Fine-Tuning with Hard Examples

• intro: ECCV 2016
• project page(paper+code+data): http://cmp.felk.cvut.cz/~radenfil/projects/siamac.html
• arxiv: https://arxiv.org/abs/1604.02426
• paper: http://cmp.felk.cvut.cz/~radenfil/publications/Radenovic-ECCV16.pdf
• code(Matlab): http://ptak.felk.cvut.cz/personal/radenfil/siamac/siaMAC_code.tar.gz

SSDH: Semi-supervised Deep Hashing for Large Scale Image Retrieval

• arxiv: http://arxiv.org/abs/1607.08477

Deep Semantic-Preserving and Ranking-Based Hashing for Image Retrieval

• intro: Microsoft
• paper: http://www.microsoft.com/en-us/research/wp-content/uploads/2016/08/Deep-Semantic-Preserving-and-Ranking-Based-Hashing-for-Image-Retrieval.pdf

PicHunt: Social Media Image Retrieval for Improved Law Enforcement

• arxiv: http://arxiv.org/abs/1608.00905

SIFT Meets CNN: A Decade Survey of Instance Retrieval

• arxiv: http://arxiv.org/abs/1608.01807

Deep Hashing: A Joint Approach for Image Signature Learning

• arxiv: http://arxiv.org/abs/1608.03658

Transitive Hashing Network for Heterogeneous Multimedia Retrieval

• intro: state of the art on NUS-WIDE, ImageNet-YahooQA
• arxiv: http://arxiv.org/abs/1608.04307

The Sketchy Database: Learning to Retrieve Badly Drawn Bunnies

• project page: http://sketchy.eye.gatech.edu/
• paper: http://www.cc.gatech.edu/~hays/tmp/sketchy-database.pdf
• github: https://github.com/janesjanes/sketchy

End-to-end Learning of Deep Visual Representations for Image Retrieval

• intro: ECCV 2016
• project page: http://www.xrce.xerox.com/Research-Development/Computer-Vision/Learning-Visual-Representations/Deep-Image-Retrieval
• arxiv: https://arxiv.org/abs/1610.07940

What Is the Best Practice for CNNs Applied to Visual Instance Retrieval?

• arxiv: https://arxiv.org/abs/1611.01640

Deep Residual Hashing

• arxiv: https://arxiv.org/abs/1612.05400

Image Retrieval with Deep Local Features and Attention-based Keypoints

• arxiv: https://arxiv.org/abs/1612.05478

Internet-Based Image Retrieval Using End-to-End Trained Deep Distributions

• arxiv: https://arxiv.org/abs/1612.07697

Compression of Deep Neural Networks for Image Instance Retrieval

• intro: DCC 2017
• arxiv: https://arxiv.org/abs/1701.04923

Effective Multi-Query Expansions: Collaborative Deep Networks for Robust Landmark Retrieval

• arxiv: https://arxiv.org/abs/1701.05003

Deep Region Hashing for Efficient Large-scale Instance Search from Images

• intro: Columbia University & University of Electronic Science and Technology of China
• arxiv: https://arxiv.org/abs/1701.07901

HashNet: Deep Learning to Hash by Continuation

• arxiv: https://arxiv.org/abs/1702.00758

Siamese Network of Deep Fisher-Vector Descriptors for Image Retrieval

• arxiv: https://arxiv.org/abs/1702.00338

Deep Geometric Retrieval

• arxiv: https://arxiv.org/abs/1702.06383

Unsupervised Triplet Hashing for Fast Image Retrieval

• arxiv: https://www.arxiv.org/abs/1702.08798

Context Aware Query Image Representation for Particular Object Retrieval

https://www.arxiv.org/abs/1703.01226

Video Indexing / Retrieval

Face Video Retrieval via Deep Learning of Binary Hash Representations

• paper: https://www.aaai.org/ocs/index.php/AAAI/AAAI16/paper/download/11893/12117

Deep Learning Based Semantic Video Indexing and Retrieval

• arxiv: https://arxiv.org/abs/1601.07754

Learning Joint Representations of Videos and Sentences with Web Image Search

• intro: 4th Workshop on Web-scale Vision and Social Media (VSM), ECCV 2016
• arxiv: http://arxiv.org/abs/1608.02367

Multi-View Product Image Search Using ConvNets Features

• arxiv: http://arxiv.org/abs/1608.03462

Generalisation and Sharing in Triplet Convnets for Sketch based Visual Search

• arxiv: https://arxiv.org/abs/1611.05301

Binary Subspace Coding for Query-by-Image Video Retrieval

• arxiv: https://arxiv.org/abs/1612.01657

Deep Supervised Hashing with Triplet Labels

• intro: ACCV 2016
• arxiv: https://arxiv.org/abs/1612.03900

Learning to Rank

Simple to Complex Cross-modal Learning to Rank

• intro: Xi’an Jiaotong University & University of Technology Sydney & National University of Singapore & CMU
• arxiv: https://arxiv.org/abs/1702.01229

Talks / Slides

TiefVision: end-to-end image similarity search engine

• intro: It covers image classification, image location ( OverFeat ) and image similarity ( Deep Ranking).
• slides:https://docs.google.com/presentation/d/16hrXJhOzkbmla9AL7JCreCuBsa5L80gm71Pfrjo7F9Y/edit#slide=id.p
• github: https://github.com/paucarre/tiefvision

Datasets

Oxford5k

Paris6k

Oxford105k

UKB

NUS-WIDE

ImageNet-YahooQA

Projects

Visual Search Server

• intro: A simple implementation of Visual Search using features extracted from Tensorflow inception model and Approximate Nearest Neighbors
• github: https://github.com/AKSHAYUBHAT/VisualSearchServer

Vehicle Retrieval: vehicle image retrieval using k CNNs ensemble method

• intro: ranked 1st and won the special prize in the final of the 3rd National Gradute Contest on Smart-CIty Technology and Creative Design, China
• project page: https://www.pkuml.org/resources/pku-vehicleid.html
• github: https://github.com/iamhankai/vehicle-retrieval-kCNNs

Blogs

Where can I buy a chair like that? – This app will tell you

• blog: http://www.news.cornell.edu/stories/2016/08/where-can-i-buy-chair-app-will-tell-you

Using Sketches to Search for Products Online

• homepage: http://sketchx.eecs.qmul.ac.uk/
• blog: https://news.developer.nvidia.com/using-sketches-to-search-for-products-online/

Tutorials

Deep Image Retrieval: Learning global representations for image search

• youtube: https://www.youtube.com/watch?v=yT52xDML6ys

Image Instance Retrieval: Overview of state-of-the-art

• youtube: https://www.youtube.com/watch?v=EYq-rpaZn1o