Cross-Modal Hamming Hashing View Full Text


Ontology type: schema:Chapter     


Chapter Info

DATE

2018-10-06

AUTHORS

Yue Cao , Bin Liu , Mingsheng Long , Jianmin Wang

ABSTRACT

Cross-modal hashing enables similarity retrieval across different content modalities, such as searching relevant images in response to text queries. It provide with the advantages of computation efficiency and retrieval quality for multimedia retrieval. Hamming space retrieval enables efficient constant-time search that returns data items within a given Hamming radius to each query, by hash lookups instead of linear scan. However, Hamming space retrieval is ineffective in existing cross-modal hashing methods, subject to their weak capability of concentrating the relevant items to be within a small Hamming ball, while worse still, the Hamming distances between hash codes from different modalities are inevitably large due to the large heterogeneity across different modalities. This work presents Cross-Modal Hamming Hashing (CMHH), a novel deep cross-modal hashing approach that generates compact and highly concentrated hash codes to enable efficient and effective Hamming space retrieval. The main idea is to penalize significantly on similar cross-modal pairs with Hamming distance larger than the Hamming radius threshold, by designing a pairwise focal loss based on the exponential distribution. Extensive experiments demonstrate that CMHH can generate highly concentrated hash codes and achieve state-of-the-art cross-modal retrieval performance for both hash lookups and linear scan scenarios on three benchmark datasets, NUS-WIDE, MIRFlickr-25K, and IAPR TC-12. More... »

PAGES

207-223

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-030-01246-5_13

DOI

http://dx.doi.org/10.1007/978-3-030-01246-5_13

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1107454624


Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
Incoming Citations Browse incoming citations for this publication using opencitations.net

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/08", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Information and Computing Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0806", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Information Systems", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "National Engineering Laboratory for Big Data Software, Beijing National Research Center for Information Science and Technology, Beijing, China", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "School of Software, Tsinghua University, Beijing, China", 
            "National Engineering Laboratory for Big Data Software, Beijing National Research Center for Information Science and Technology, Beijing, China"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Cao", 
        "givenName": "Yue", 
        "id": "sg:person.012712534041.97", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012712534041.97"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "National Engineering Laboratory for Big Data Software, Beijing National Research Center for Information Science and Technology, Beijing, China", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "School of Software, Tsinghua University, Beijing, China", 
            "National Engineering Laboratory for Big Data Software, Beijing National Research Center for Information Science and Technology, Beijing, China"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Liu", 
        "givenName": "Bin", 
        "id": "sg:person.011667021334.12", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011667021334.12"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "National Engineering Laboratory for Big Data Software, Beijing National Research Center for Information Science and Technology, Beijing, China", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "School of Software, Tsinghua University, Beijing, China", 
            "National Engineering Laboratory for Big Data Software, Beijing National Research Center for Information Science and Technology, Beijing, China"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Long", 
        "givenName": "Mingsheng", 
        "id": "sg:person.013417115303.81", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013417115303.81"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "National Engineering Laboratory for Big Data Software, Beijing National Research Center for Information Science and Technology, Beijing, China", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "School of Software, Tsinghua University, Beijing, China", 
            "National Engineering Laboratory for Big Data Software, Beijing National Research Center for Information Science and Technology, Beijing, China"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Wang", 
        "givenName": "Jianmin", 
        "id": "sg:person.012303351315.43", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012303351315.43"
        ], 
        "type": "Person"
      }
    ], 
    "datePublished": "2018-10-06", 
    "datePublishedReg": "2018-10-06", 
    "description": "Cross-modal hashing enables similarity retrieval across different content modalities, such as searching relevant images in response to text queries. It provide with the advantages of computation efficiency and retrieval quality for multimedia retrieval. Hamming space retrieval enables efficient constant-time search that returns data items within a given Hamming radius to each query, by hash lookups instead of linear scan. However, Hamming space retrieval is ineffective in existing cross-modal hashing methods, subject to their weak capability of concentrating the relevant items to be within a small Hamming ball, while worse still, the Hamming distances between hash codes from different modalities are inevitably large due to the large heterogeneity across different modalities. This work presents Cross-Modal Hamming Hashing (CMHH), a novel deep cross-modal hashing approach that generates compact and highly concentrated hash codes to enable efficient and effective Hamming space retrieval. The main idea is to penalize significantly on similar cross-modal pairs with Hamming distance larger than the Hamming radius threshold, by designing a pairwise focal loss based on the exponential distribution. Extensive experiments demonstrate that CMHH can generate highly concentrated hash codes and achieve state-of-the-art cross-modal retrieval performance for both hash lookups and linear scan scenarios on three benchmark datasets, NUS-WIDE, MIRFlickr-25K, and IAPR TC-12.", 
    "editor": [
      {
        "familyName": "Ferrari", 
        "givenName": "Vittorio", 
        "type": "Person"
      }, 
      {
        "familyName": "Hebert", 
        "givenName": "Martial", 
        "type": "Person"
      }, 
      {
        "familyName": "Sminchisescu", 
        "givenName": "Cristian", 
        "type": "Person"
      }, 
      {
        "familyName": "Weiss", 
        "givenName": "Yair", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-030-01246-5_13", 
    "inLanguage": "en", 
    "isAccessibleForFree": false, 
    "isPartOf": {
      "isbn": [
        "978-3-030-01245-8", 
        "978-3-030-01246-5"
      ], 
      "name": "Computer Vision \u2013 ECCV 2018", 
      "type": "Book"
    }, 
    "keywords": [
      "hash codes", 
      "space retrieval", 
      "hash lookups", 
      "cross-modal retrieval performance", 
      "cross-modal hashing methods", 
      "NUS-WIDE", 
      "Hamming distance", 
      "multimedia retrieval", 
      "text queries", 
      "IAPR TC-12", 
      "similarity retrieval", 
      "relevant images", 
      "hashing methods", 
      "different modalities", 
      "data items", 
      "Extensive experiments", 
      "retrieval performance", 
      "retrieval quality", 
      "benchmark datasets", 
      "content modalities", 
      "MIRFLICKR-25K", 
      "radius threshold", 
      "computation efficiency", 
      "scan scenarios", 
      "retrieval", 
      "main idea", 
      "queries", 
      "relevant items", 
      "lookup", 
      "weak capability", 
      "linear scan", 
      "code", 
      "Hamming ball", 
      "datasets", 
      "Hamming", 
      "focal loss", 
      "images", 
      "scenarios", 
      "capability", 
      "search", 
      "performance", 
      "advantages", 
      "items", 
      "idea", 
      "distance", 
      "efficiency", 
      "quality", 
      "work", 
      "method", 
      "experiments", 
      "modalities", 
      "exponential distribution", 
      "threshold", 
      "state", 
      "pairs", 
      "scans", 
      "ball", 
      "heterogeneity", 
      "distribution", 
      "loss", 
      "large heterogeneity", 
      "radius", 
      "response", 
      "approach", 
      "different content modalities", 
      "efficient constant-time search", 
      "constant-time search", 
      "Hamming radius", 
      "Hamming space retrieval", 
      "small Hamming ball", 
      "Cross-Modal Hamming", 
      "concentrated hash codes", 
      "effective Hamming space retrieval", 
      "similar cross-modal pairs", 
      "cross-modal pairs", 
      "Hamming radius threshold", 
      "pairwise focal loss", 
      "CMHH", 
      "art cross-modal retrieval performance", 
      "TC-12"
    ], 
    "name": "Cross-Modal Hamming Hashing", 
    "pagination": "207-223", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1107454624"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-030-01246-5_13"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-030-01246-5_13", 
      "https://app.dimensions.ai/details/publication/pub.1107454624"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2022-01-01T19:06", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20220101/entities/gbq_results/chapter/chapter_117.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-030-01246-5_13"
  }
]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-01246-5_13'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-01246-5_13'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-01246-5_13'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-01246-5_13'


 

This table displays all metadata directly associated to this object as RDF triples.

177 TRIPLES      23 PREDICATES      105 URIs      98 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-030-01246-5_13 schema:about anzsrc-for:08
2 anzsrc-for:0806
3 schema:author Ncc8491107a0a463f86d891a94dd5164a
4 schema:datePublished 2018-10-06
5 schema:datePublishedReg 2018-10-06
6 schema:description Cross-modal hashing enables similarity retrieval across different content modalities, such as searching relevant images in response to text queries. It provide with the advantages of computation efficiency and retrieval quality for multimedia retrieval. Hamming space retrieval enables efficient constant-time search that returns data items within a given Hamming radius to each query, by hash lookups instead of linear scan. However, Hamming space retrieval is ineffective in existing cross-modal hashing methods, subject to their weak capability of concentrating the relevant items to be within a small Hamming ball, while worse still, the Hamming distances between hash codes from different modalities are inevitably large due to the large heterogeneity across different modalities. This work presents Cross-Modal Hamming Hashing (CMHH), a novel deep cross-modal hashing approach that generates compact and highly concentrated hash codes to enable efficient and effective Hamming space retrieval. The main idea is to penalize significantly on similar cross-modal pairs with Hamming distance larger than the Hamming radius threshold, by designing a pairwise focal loss based on the exponential distribution. Extensive experiments demonstrate that CMHH can generate highly concentrated hash codes and achieve state-of-the-art cross-modal retrieval performance for both hash lookups and linear scan scenarios on three benchmark datasets, NUS-WIDE, MIRFlickr-25K, and IAPR TC-12.
7 schema:editor N0a00459af880439dae22055a6b2a9d94
8 schema:genre chapter
9 schema:inLanguage en
10 schema:isAccessibleForFree false
11 schema:isPartOf Nd8def8c4e98d4240a183b5dc4646fe37
12 schema:keywords CMHH
13 Cross-Modal Hamming
14 Extensive experiments
15 Hamming
16 Hamming ball
17 Hamming distance
18 Hamming radius
19 Hamming radius threshold
20 Hamming space retrieval
21 IAPR TC-12
22 MIRFLICKR-25K
23 NUS-WIDE
24 TC-12
25 advantages
26 approach
27 art cross-modal retrieval performance
28 ball
29 benchmark datasets
30 capability
31 code
32 computation efficiency
33 concentrated hash codes
34 constant-time search
35 content modalities
36 cross-modal hashing methods
37 cross-modal pairs
38 cross-modal retrieval performance
39 data items
40 datasets
41 different content modalities
42 different modalities
43 distance
44 distribution
45 effective Hamming space retrieval
46 efficiency
47 efficient constant-time search
48 experiments
49 exponential distribution
50 focal loss
51 hash codes
52 hash lookups
53 hashing methods
54 heterogeneity
55 idea
56 images
57 items
58 large heterogeneity
59 linear scan
60 lookup
61 loss
62 main idea
63 method
64 modalities
65 multimedia retrieval
66 pairs
67 pairwise focal loss
68 performance
69 quality
70 queries
71 radius
72 radius threshold
73 relevant images
74 relevant items
75 response
76 retrieval
77 retrieval performance
78 retrieval quality
79 scan scenarios
80 scans
81 scenarios
82 search
83 similar cross-modal pairs
84 similarity retrieval
85 small Hamming ball
86 space retrieval
87 state
88 text queries
89 threshold
90 weak capability
91 work
92 schema:name Cross-Modal Hamming Hashing
93 schema:pagination 207-223
94 schema:productId N7b295d521a7345df8e6d760cc5967c2e
95 Neea4d800e50045d8ac4dc8702326679d
96 schema:publisher N23027c9be7784f14870c828939dca749
97 schema:sameAs https://app.dimensions.ai/details/publication/pub.1107454624
98 https://doi.org/10.1007/978-3-030-01246-5_13
99 schema:sdDatePublished 2022-01-01T19:06
100 schema:sdLicense https://scigraph.springernature.com/explorer/license/
101 schema:sdPublisher Nb98932c5a87f49a1ba18c8e487c1ca66
102 schema:url https://doi.org/10.1007/978-3-030-01246-5_13
103 sgo:license sg:explorer/license/
104 sgo:sdDataset chapters
105 rdf:type schema:Chapter
106 N0a00459af880439dae22055a6b2a9d94 rdf:first N1cf41002ef144d78adb87fb5137bf062
107 rdf:rest N608e3e3f8446480e86649fc96013d3ff
108 N1cf41002ef144d78adb87fb5137bf062 schema:familyName Ferrari
109 schema:givenName Vittorio
110 rdf:type schema:Person
111 N23027c9be7784f14870c828939dca749 schema:name Springer Nature
112 rdf:type schema:Organisation
113 N4fcb59b7c3544331bb9deee42eb52bf4 rdf:first sg:person.011667021334.12
114 rdf:rest N82561473f5be470893af7787ec182be3
115 N5c2bb85439cd44d2bdde8fa1ab1d8def rdf:first sg:person.012303351315.43
116 rdf:rest rdf:nil
117 N608e3e3f8446480e86649fc96013d3ff rdf:first N8e573fb9c5d247f9867709cc4b6f1a28
118 rdf:rest Na60c825305f54af08ff0fac5142c59d8
119 N7b295d521a7345df8e6d760cc5967c2e schema:name doi
120 schema:value 10.1007/978-3-030-01246-5_13
121 rdf:type schema:PropertyValue
122 N82561473f5be470893af7787ec182be3 rdf:first sg:person.013417115303.81
123 rdf:rest N5c2bb85439cd44d2bdde8fa1ab1d8def
124 N8e573fb9c5d247f9867709cc4b6f1a28 schema:familyName Hebert
125 schema:givenName Martial
126 rdf:type schema:Person
127 Na60c825305f54af08ff0fac5142c59d8 rdf:first Nd6d8efbe03a84ab49ac84de7594b96ef
128 rdf:rest Nf5fd33a2e84b48ef8976825b1ca3b8d7
129 Nb98932c5a87f49a1ba18c8e487c1ca66 schema:name Springer Nature - SN SciGraph project
130 rdf:type schema:Organization
131 Ncc8491107a0a463f86d891a94dd5164a rdf:first sg:person.012712534041.97
132 rdf:rest N4fcb59b7c3544331bb9deee42eb52bf4
133 Nd6d8efbe03a84ab49ac84de7594b96ef schema:familyName Sminchisescu
134 schema:givenName Cristian
135 rdf:type schema:Person
136 Nd8def8c4e98d4240a183b5dc4646fe37 schema:isbn 978-3-030-01245-8
137 978-3-030-01246-5
138 schema:name Computer Vision – ECCV 2018
139 rdf:type schema:Book
140 Neea4d800e50045d8ac4dc8702326679d schema:name dimensions_id
141 schema:value pub.1107454624
142 rdf:type schema:PropertyValue
143 Nf1b74e5054054074adb838e41b25100e schema:familyName Weiss
144 schema:givenName Yair
145 rdf:type schema:Person
146 Nf5fd33a2e84b48ef8976825b1ca3b8d7 rdf:first Nf1b74e5054054074adb838e41b25100e
147 rdf:rest rdf:nil
148 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
149 schema:name Information and Computing Sciences
150 rdf:type schema:DefinedTerm
151 anzsrc-for:0806 schema:inDefinedTermSet anzsrc-for:
152 schema:name Information Systems
153 rdf:type schema:DefinedTerm
154 sg:person.011667021334.12 schema:affiliation grid-institutes:None
155 schema:familyName Liu
156 schema:givenName Bin
157 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011667021334.12
158 rdf:type schema:Person
159 sg:person.012303351315.43 schema:affiliation grid-institutes:None
160 schema:familyName Wang
161 schema:givenName Jianmin
162 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012303351315.43
163 rdf:type schema:Person
164 sg:person.012712534041.97 schema:affiliation grid-institutes:None
165 schema:familyName Cao
166 schema:givenName Yue
167 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012712534041.97
168 rdf:type schema:Person
169 sg:person.013417115303.81 schema:affiliation grid-institutes:None
170 schema:familyName Long
171 schema:givenName Mingsheng
172 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013417115303.81
173 rdf:type schema:Person
174 grid-institutes:None schema:alternateName National Engineering Laboratory for Big Data Software, Beijing National Research Center for Information Science and Technology, Beijing, China
175 schema:name National Engineering Laboratory for Big Data Software, Beijing National Research Center for Information Science and Technology, Beijing, China
176 School of Software, Tsinghua University, Beijing, China
177 rdf:type schema:Organization
 




Preview window. Press ESC to close (or click here)


...