Uncertainty Estimates and Multi-hypotheses Networks for Optical Flow View Full Text


Ontology type: schema:Chapter      Open Access: True


Chapter Info

DATE

2018-10-06

AUTHORS

Eddy Ilg , Özgün Çiçek , Silvio Galesso , Aaron Klein , Osama Makansi , Frank Hutter , Thomas Brox

ABSTRACT

Optical flow estimation can be formulated as an end-to-end supervised learning problem, which yields estimates with a superior accuracy-runtime tradeoff compared to alternative methodology. In this paper, we make such networks estimate their local uncertainty about the correctness of their prediction, which is vital information when building decisions on top of the estimations. For the first time we compare several strategies and techniques to estimate uncertainty in a large-scale computer vision task like optical flow estimation. Moreover, we introduce a new network architecture and loss function that enforce complementary hypotheses and provide uncertainty estimates efficiently with a single forward pass and without the need for sampling or ensembles. We demonstrate the quality of the uncertainty estimates, which is clearly above previous confidence measures on optical flow and allows for interactive frame rates. More... »

PAGES

677-693

Book

TITLE

Computer Vision – ECCV 2018

ISBN

978-3-030-01233-5
978-3-030-01234-2

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-030-01234-2_40

DOI

http://dx.doi.org/10.1007/978-3-030-01234-2_40

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1107454605


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/2008/0801", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Artificial Intelligence and Image Processing", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "University of Freiburg, Freiburg, Germany", 
          "id": "http://www.grid.ac/institutes/grid.5963.9", 
          "name": [
            "University of Freiburg, Freiburg, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Ilg", 
        "givenName": "Eddy", 
        "id": "sg:person.014016531047.11", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014016531047.11"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "University of Freiburg, Freiburg, Germany", 
          "id": "http://www.grid.ac/institutes/grid.5963.9", 
          "name": [
            "University of Freiburg, Freiburg, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "\u00c7i\u00e7ek", 
        "givenName": "\u00d6zg\u00fcn", 
        "id": "sg:person.016314276446.01", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016314276446.01"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "University of Freiburg, Freiburg, Germany", 
          "id": "http://www.grid.ac/institutes/grid.5963.9", 
          "name": [
            "University of Freiburg, Freiburg, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Galesso", 
        "givenName": "Silvio", 
        "id": "sg:person.015466411010.75", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015466411010.75"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "University of Freiburg, Freiburg, Germany", 
          "id": "http://www.grid.ac/institutes/grid.5963.9", 
          "name": [
            "University of Freiburg, Freiburg, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Klein", 
        "givenName": "Aaron", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "University of Freiburg, Freiburg, Germany", 
          "id": "http://www.grid.ac/institutes/grid.5963.9", 
          "name": [
            "University of Freiburg, Freiburg, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Makansi", 
        "givenName": "Osama", 
        "id": "sg:person.013221150447.86", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013221150447.86"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "University of Freiburg, Freiburg, Germany", 
          "id": "http://www.grid.ac/institutes/grid.5963.9", 
          "name": [
            "University of Freiburg, Freiburg, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Hutter", 
        "givenName": "Frank", 
        "id": "sg:person.016710317431.52", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016710317431.52"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "University of Freiburg, Freiburg, Germany", 
          "id": "http://www.grid.ac/institutes/grid.5963.9", 
          "name": [
            "University of Freiburg, Freiburg, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Brox", 
        "givenName": "Thomas", 
        "id": "sg:person.012443225372.65", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012443225372.65"
        ], 
        "type": "Person"
      }
    ], 
    "datePublished": "2018-10-06", 
    "datePublishedReg": "2018-10-06", 
    "description": "Optical flow estimation can be formulated as an end-to-end supervised learning problem, which yields estimates with a superior accuracy-runtime tradeoff compared to alternative methodology. In this paper, we make such networks estimate their local uncertainty about the correctness of their prediction, which is vital information when building decisions on top of the estimations. For the first time we compare several strategies and techniques to estimate uncertainty in a large-scale computer vision task like optical flow estimation. Moreover, we introduce a new network architecture and loss function that enforce complementary hypotheses and provide uncertainty estimates efficiently with a single forward pass and without the need for sampling or ensembles. We demonstrate the quality of the uncertainty estimates, which is clearly above previous confidence measures on optical flow and allows for interactive frame rates.", 
    "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.10.1007/978-3-030-01234-2_40", 
    "isAccessibleForFree": true, 
    "isPartOf": {
      "isbn": [
        "978-3-030-01233-5", 
        "978-3-030-01234-2"
      ], 
      "name": "Computer Vision \u2013 ECCV 2018", 
      "type": "Book"
    }, 
    "keywords": [
      "optical flow estimation", 
      "optical flow", 
      "computer vision tasks", 
      "single forward pass", 
      "interactive frame rates", 
      "supervised learning problem", 
      "new network architecture", 
      "flow estimation", 
      "vision tasks", 
      "network architecture", 
      "forward pass", 
      "learning problem", 
      "uncertainty estimates", 
      "such networks", 
      "confidence measure", 
      "frame rate", 
      "loss function", 
      "network", 
      "vital information", 
      "local uncertainty", 
      "architecture", 
      "correctness", 
      "task", 
      "estimation", 
      "Multi", 
      "tradeoff", 
      "information", 
      "uncertainty", 
      "alternative methodology", 
      "decisions", 
      "methodology", 
      "ensemble", 
      "technique", 
      "quality", 
      "prediction", 
      "top", 
      "need", 
      "pass", 
      "time", 
      "strategies", 
      "end", 
      "flow", 
      "sampling", 
      "estimates", 
      "measures", 
      "function", 
      "complementary hypothesis", 
      "first time", 
      "rate", 
      "hypothesis", 
      "paper", 
      "problem"
    ], 
    "name": "Uncertainty Estimates and Multi-hypotheses Networks for Optical Flow", 
    "pagination": "677-693", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1107454605"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-030-01234-2_40"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-030-01234-2_40", 
      "https://app.dimensions.ai/details/publication/pub.1107454605"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2022-10-01T06:55", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20221001/entities/gbq_results/chapter/chapter_25.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-030-01234-2_40"
  }
]
 

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.10.1007/978-3-030-01234-2_40'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-01234-2_40'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-01234-2_40'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-01234-2_40'


 

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

167 TRIPLES      22 PREDICATES      76 URIs      69 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-030-01234-2_40 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author N19f3a41fa65e493686d7bf17dc201a17
4 schema:datePublished 2018-10-06
5 schema:datePublishedReg 2018-10-06
6 schema:description Optical flow estimation can be formulated as an end-to-end supervised learning problem, which yields estimates with a superior accuracy-runtime tradeoff compared to alternative methodology. In this paper, we make such networks estimate their local uncertainty about the correctness of their prediction, which is vital information when building decisions on top of the estimations. For the first time we compare several strategies and techniques to estimate uncertainty in a large-scale computer vision task like optical flow estimation. Moreover, we introduce a new network architecture and loss function that enforce complementary hypotheses and provide uncertainty estimates efficiently with a single forward pass and without the need for sampling or ensembles. We demonstrate the quality of the uncertainty estimates, which is clearly above previous confidence measures on optical flow and allows for interactive frame rates.
7 schema:editor N9d8d1479d45c41e284363e8b83d3065a
8 schema:genre chapter
9 schema:isAccessibleForFree true
10 schema:isPartOf Ndf775936ccf149c8baa9309235686d10
11 schema:keywords Multi
12 alternative methodology
13 architecture
14 complementary hypothesis
15 computer vision tasks
16 confidence measure
17 correctness
18 decisions
19 end
20 ensemble
21 estimates
22 estimation
23 first time
24 flow
25 flow estimation
26 forward pass
27 frame rate
28 function
29 hypothesis
30 information
31 interactive frame rates
32 learning problem
33 local uncertainty
34 loss function
35 measures
36 methodology
37 need
38 network
39 network architecture
40 new network architecture
41 optical flow
42 optical flow estimation
43 paper
44 pass
45 prediction
46 problem
47 quality
48 rate
49 sampling
50 single forward pass
51 strategies
52 such networks
53 supervised learning problem
54 task
55 technique
56 time
57 top
58 tradeoff
59 uncertainty
60 uncertainty estimates
61 vision tasks
62 vital information
63 schema:name Uncertainty Estimates and Multi-hypotheses Networks for Optical Flow
64 schema:pagination 677-693
65 schema:productId Nb7e6c64cca894f9e8e6879ac5679f88e
66 Ncb535d692fc34566a8b0a63307eba16e
67 schema:publisher N9515048f918f4d06b03c4ba6f9428725
68 schema:sameAs https://app.dimensions.ai/details/publication/pub.1107454605
69 https://doi.org/10.1007/978-3-030-01234-2_40
70 schema:sdDatePublished 2022-10-01T06:55
71 schema:sdLicense https://scigraph.springernature.com/explorer/license/
72 schema:sdPublisher Nf986861ad72b40048de5c3e5c276a17a
73 schema:url https://doi.org/10.1007/978-3-030-01234-2_40
74 sgo:license sg:explorer/license/
75 sgo:sdDataset chapters
76 rdf:type schema:Chapter
77 N0c1f78eaded548a78a59d87a4f202851 schema:familyName Ferrari
78 schema:givenName Vittorio
79 rdf:type schema:Person
80 N174e178912914010afefb02e9c8e3680 schema:familyName Hebert
81 schema:givenName Martial
82 rdf:type schema:Person
83 N19f3a41fa65e493686d7bf17dc201a17 rdf:first sg:person.014016531047.11
84 rdf:rest N85c31d35c5034fa8bb19cfe0e5f7788f
85 N3137e7a0d70042ae8cff4a05c89d03f4 schema:familyName Weiss
86 schema:givenName Yair
87 rdf:type schema:Person
88 N5c3e01053e24453eb34e9597e32f6c9c schema:affiliation grid-institutes:grid.5963.9
89 schema:familyName Klein
90 schema:givenName Aaron
91 rdf:type schema:Person
92 N7c9cd8e77adc4a169821a256e899b4fd rdf:first sg:person.013221150447.86
93 rdf:rest Nacbe15fadc944a1f89ef42ae9e72f136
94 N85c31d35c5034fa8bb19cfe0e5f7788f rdf:first sg:person.016314276446.01
95 rdf:rest Ne5643c474e7d4be386aceabc390a4acf
96 N899b09370ab944efacac5c3dbb24cc1e rdf:first N3137e7a0d70042ae8cff4a05c89d03f4
97 rdf:rest rdf:nil
98 N8e29ce8bed4643e69a64a17eadf55406 rdf:first sg:person.012443225372.65
99 rdf:rest rdf:nil
100 N9515048f918f4d06b03c4ba6f9428725 schema:name Springer Nature
101 rdf:type schema:Organisation
102 N9d8d1479d45c41e284363e8b83d3065a rdf:first N0c1f78eaded548a78a59d87a4f202851
103 rdf:rest Nd6e81d0626de4a11bd5e482472b12ba0
104 Na8fd92d4013842ecac702b5d3c2f8d14 schema:familyName Sminchisescu
105 schema:givenName Cristian
106 rdf:type schema:Person
107 Nacbe15fadc944a1f89ef42ae9e72f136 rdf:first sg:person.016710317431.52
108 rdf:rest N8e29ce8bed4643e69a64a17eadf55406
109 Nb7e6c64cca894f9e8e6879ac5679f88e schema:name doi
110 schema:value 10.1007/978-3-030-01234-2_40
111 rdf:type schema:PropertyValue
112 Nbd29fbe4cc9244e89d4e18491fe42520 rdf:first N5c3e01053e24453eb34e9597e32f6c9c
113 rdf:rest N7c9cd8e77adc4a169821a256e899b4fd
114 Nc6d7a9f718444c139386bf73cb1b0cc1 rdf:first Na8fd92d4013842ecac702b5d3c2f8d14
115 rdf:rest N899b09370ab944efacac5c3dbb24cc1e
116 Ncb535d692fc34566a8b0a63307eba16e schema:name dimensions_id
117 schema:value pub.1107454605
118 rdf:type schema:PropertyValue
119 Nd6e81d0626de4a11bd5e482472b12ba0 rdf:first N174e178912914010afefb02e9c8e3680
120 rdf:rest Nc6d7a9f718444c139386bf73cb1b0cc1
121 Ndf775936ccf149c8baa9309235686d10 schema:isbn 978-3-030-01233-5
122 978-3-030-01234-2
123 schema:name Computer Vision – ECCV 2018
124 rdf:type schema:Book
125 Ne5643c474e7d4be386aceabc390a4acf rdf:first sg:person.015466411010.75
126 rdf:rest Nbd29fbe4cc9244e89d4e18491fe42520
127 Nf986861ad72b40048de5c3e5c276a17a schema:name Springer Nature - SN SciGraph project
128 rdf:type schema:Organization
129 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
130 schema:name Information and Computing Sciences
131 rdf:type schema:DefinedTerm
132 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
133 schema:name Artificial Intelligence and Image Processing
134 rdf:type schema:DefinedTerm
135 sg:person.012443225372.65 schema:affiliation grid-institutes:grid.5963.9
136 schema:familyName Brox
137 schema:givenName Thomas
138 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012443225372.65
139 rdf:type schema:Person
140 sg:person.013221150447.86 schema:affiliation grid-institutes:grid.5963.9
141 schema:familyName Makansi
142 schema:givenName Osama
143 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013221150447.86
144 rdf:type schema:Person
145 sg:person.014016531047.11 schema:affiliation grid-institutes:grid.5963.9
146 schema:familyName Ilg
147 schema:givenName Eddy
148 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014016531047.11
149 rdf:type schema:Person
150 sg:person.015466411010.75 schema:affiliation grid-institutes:grid.5963.9
151 schema:familyName Galesso
152 schema:givenName Silvio
153 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015466411010.75
154 rdf:type schema:Person
155 sg:person.016314276446.01 schema:affiliation grid-institutes:grid.5963.9
156 schema:familyName Çiçek
157 schema:givenName Özgün
158 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016314276446.01
159 rdf:type schema:Person
160 sg:person.016710317431.52 schema:affiliation grid-institutes:grid.5963.9
161 schema:familyName Hutter
162 schema:givenName Frank
163 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016710317431.52
164 rdf:type schema:Person
165 grid-institutes:grid.5963.9 schema:alternateName University of Freiburg, Freiburg, Germany
166 schema:name University of Freiburg, Freiburg, Germany
167 rdf:type schema:Organization
 




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


...