{"data":{"id":"10.4121/dc5e2fb8-6005-40cd-9afa-ff03c57d0a23","type":"dois","attributes":{"doi":"10.4121/dc5e2fb8-6005-40cd-9afa-ff03c57d0a23","prefix":"10.4121","suffix":"dc5e2fb8-6005-40cd-9afa-ff03c57d0a23","identifiers":[],"alternateIdentifiers":[],"creators":[{"name":"Strafforello, Ombretta","givenName":"Ombretta","familyName":"Strafforello","affiliation":[],"nameIdentifiers":[]},{"name":"Liu, Xin","givenName":"Xin","familyName":"Liu","affiliation":[],"nameIdentifiers":[]},{"name":"van Gemert, Jan","nameType":"Personal","givenName":"Jan","familyName":"van Gemert","nameIdentifiers":[{"nameIdentifier":"0000-0002-6913-0482","nameIdentifierScheme":"https://orcid.org/"}],"affiliation":[]},{"name":"Schutte , Klamer","nameType":"Personal","givenName":"Klamer","familyName":"Schutte ","nameIdentifiers":[{"nameIdentifier":"0000-0002-9954-0685","nameIdentifierScheme":"https://orcid.org/"}],"affiliation":[]}],"titles":[{"title":"Code underlying the publication: \"Video BagNet: short temporal receptive fields increase robustness in long-term action recognition\""}],"publisher":"4TU.ResearchData","container":{},"publicationYear":2024,"subjects":[{"subject":"Artificial Intelligence and Image Processing","subjectScheme":"Australian and New Zealand Standard Research Classification (ANZSRC), 2008","classificationCode":"0801"},{"subject":"FOS: Computer and information sciences","schemeUri":"http://www.oecd.org/science/inno/38235147.pdf","subjectScheme":"Fields of Science and Technology (FOS)"},{"subject":"Information and Computing Sciences","subjectScheme":"Australian and New Zealand Standard Research Classification (ANZSRC), 2008","classificationCode":"08"},{"subject":"Computer vision"},{"subject":"Action recognition"},{"subject":"3D-CNN"},{"subject":"BagNet"},{"subject":"Temporal receptive field"}],"contributors":[{"name":"TU Delft, TU Delft, Faculty Of Electrical Engineering, Mathematics And Computer Science, Department Of Intelligent Systems, Computer Vision Lab","nameType":"Organizational","affiliation":[],"contributorType":"Other","nameIdentifiers":[]},{"name":"TNO, Netherlands Organisation For Applied Scientific Research, Intelligent Imaging Group","nameType":"Organizational","affiliation":[],"contributorType":"Other","nameIdentifiers":[]}],"dates":[{"date":"2024-05-24","dateType":"Issued"}],"language":"en","types":{"ris":"COMP","bibtex":"misc","citeproc":"article","schemaOrg":"SoftwareSourceCode","resourceType":"Software","resourceTypeGeneral":"Software"},"relatedIdentifiers":[{"relationType":"IsSupplementTo","relatedIdentifier":"10.1109/iccvw60793.2023.00023","relatedIdentifierType":"DOI"}],"relatedItems":[],"sizes":[],"formats":["GitHub repository with python code"],"version":"1","rightsList":[{"rights":"Creative Commons Zero v1.0 Universal","rightsUri":"https://creativecommons.org/publicdomain/zero/1.0/legalcode","schemeUri":"https://spdx.org/licenses/","rightsIdentifier":"cc0-1.0","rightsIdentifierScheme":"SPDX"}],"descriptions":[{"description":"Previous work on long-term video action recognition relies on deep 3D-convolutional models that have a large temporal receptive field (RF). We argue that these models are not always the best choice for temporal modeling in videos. A large temporal receptive field allows the model to encode the exact sub-action order of a video, which causes a performance decrease when testing videos have a different sub-action order. In this work, we investigate whether we can improve the model robustness to the sub-action order by shrinking the temporal receptive field of action recognition models. For this, we design Video BagNet, a variant of the 3D ResNet-50 model with the temporal receptive field size limited to 1, 9, 17 or 33 frames. We analyze Video Bag-Net on synthetic and real-world video datasets and experimentally compare models with varying temporal receptive fields. We find that short receptive fields are robust to sub-action order changes, while larger temporal receptive fields are sensitive to the sub-action order. In this repository, we provide our code, including the implementation of Video Bag-Net.","descriptionType":"Abstract"}],"geoLocations":[],"fundingReferences":[],"xml":"PD94bWwgdmVyc2lvbj0iMS4wIiBlbmNvZGluZz0iVVRGLTgiPz4KPHJlc291cmNlIHhtbG5zPSJodHRwOi8vZGF0YWNpdGUub3JnL3NjaGVtYS9rZXJuZWwtNCIgeG1sbnM6eHNpPSJodHRwOi8vd3d3LnczLm9yZy8yMDAxL1hNTFNjaGVtYS1pbnN0YW5jZSIgeHNpOnNjaGVtYUxvY2F0aW9uPSJodHRwOi8vZGF0YWNpdGUub3JnL3NjaGVtYS9rZXJuZWwtNCBodHRwOi8vc2NoZW1hLmRhdGFjaXRlLm9yZy9tZXRhL2tlcm5lbC00LjQvbWV0YWRhdGEueHNkIj4KICA8aWRlbnRpZmllciBpZGVudGlmaWVyVHlwZT0iRE9JIj4xMC40MTIxL0RDNUUyRkI4LTYwMDUtNDBDRC05QUZBLUZGMDNDNTdEMEEyMzwvaWRlbnRpZmllcj4KICA8Y3JlYXRvcnM+CiAgICA8Y3JlYXRvcj4KICAgICAgPGNyZWF0b3JOYW1lPk9tYnJldHRhIFN0cmFmZm9yZWxsbzwvY3JlYXRvck5hbWU+CiAgICAgIDxnaXZlbk5hbWU+T21icmV0dGE8L2dpdmVuTmFtZT4KICAgICAgPGZhbWlseU5hbWU+U3RyYWZmb3JlbGxvPC9mYW1pbHlOYW1lPgogICAgPC9jcmVhdG9yPgogICAgPGNyZWF0b3I+CiAgICAgIDxjcmVhdG9yTmFtZT5YaW4gTGl1PC9jcmVhdG9yTmFtZT4KICAgICAgPGdpdmVuTmFtZT5YaW48L2dpdmVuTmFtZT4KICAgICAgPGZhbWlseU5hbWU+TGl1PC9mYW1pbHlOYW1lPgogICAgPC9jcmVhdG9yPgogICAgPGNyZWF0b3I+CiAgICAgIDxjcmVhdG9yTmFtZSBuYW1lVHlwZT0iUGVyc29uYWwiPkphbiB2YW4gR2VtZXJ0PC9jcmVhdG9yTmFtZT4KICAgICAgPGdpdmVuTmFtZT5KYW48L2dpdmVuTmFtZT4KICAgICAgPGZhbWlseU5hbWU+dmFuIEdlbWVydDwvZmFtaWx5TmFtZT4KICAgICAgPG5hbWVJZGVudGlmaWVyIG5hbWVJZGVudGlmaWVyU2NoZW1lPSJodHRwczovL29yY2lkLm9yZy8iPjAwMDAtMDAwMi02OTEzLTA0ODI8L25hbWVJZGVudGlmaWVyPgogICAgPC9jcmVhdG9yPgogICAgPGNyZWF0b3I+CiAgICAgIDxjcmVhdG9yTmFtZSBuYW1lVHlwZT0iUGVyc29uYWwiPktsYW1lciBTY2h1dHRlIDwvY3JlYXRvck5hbWU+CiAgICAgIDxnaXZlbk5hbWU+S2xhbWVyPC9naXZlbk5hbWU+CiAgICAgIDxmYW1pbHlOYW1lPlNjaHV0dGUgPC9mYW1pbHlOYW1lPgogICAgICA8bmFtZUlkZW50aWZpZXIgbmFtZUlkZW50aWZpZXJTY2hlbWU9Imh0dHBzOi8vb3JjaWQub3JnLyI+MDAwMC0wMDAyLTk5NTQtMDY4NTwvbmFtZUlkZW50aWZpZXI+CiAgICA8L2NyZWF0b3I+CiAgPC9jcmVhdG9ycz4KICA8dGl0bGVzPgogICAgPHRpdGxlPkNvZGUgdW5kZXJseWluZyB0aGUgcHVibGljYXRpb246ICJWaWRlbyBCYWdOZXQ6IHNob3J0IHRlbXBvcmFsIHJlY2VwdGl2ZSBmaWVsZHMgaW5jcmVhc2Ugcm9idXN0bmVzcyBpbiBsb25nLXRlcm0gYWN0aW9uIHJlY29nbml0aW9uIjwvdGl0bGU+CiAgPC90aXRsZXM+CiAgPHB1Ymxpc2hlcj40VFUuUmVzZWFyY2hEYXRhPC9wdWJsaXNoZXI+CiAgPHB1YmxpY2F0aW9uWWVhcj4yMDI0PC9wdWJsaWNhdGlvblllYXI+CiAgPHJlc291cmNlVHlwZSByZXNvdXJjZVR5cGVHZW5lcmFsPSJTb2Z0d2FyZSI+U29mdHdhcmU8L3Jlc291cmNlVHlwZT4KICA8c3ViamVjdHM+CiAgICA8c3ViamVjdCBzdWJqZWN0U2NoZW1lPSJBdXN0cmFsaWFuIGFuZCBOZXcgWmVhbGFuZCBTdGFuZGFyZCBSZXNlYXJjaCBDbGFzc2lmaWNhdGlvbiAoQU5aU1JDKSwgMjAwOCIgY2xhc3NpZmljYXRpb25Db2RlPSIwODAxIj5BcnRpZmljaWFsIEludGVsbGlnZW5jZSBhbmQgSW1hZ2UgUHJvY2Vzc2luZzwvc3ViamVjdD4KICAgIDxzdWJqZWN0IHN1YmplY3RTY2hlbWU9IkF1c3RyYWxpYW4gYW5kIE5ldyBaZWFsYW5kIFN0YW5kYXJkIFJlc2VhcmNoIENsYXNzaWZpY2F0aW9uIChBTlpTUkMpLCAyMDA4IiBjbGFzc2lmaWNhdGlvbkNvZGU9IjA4Ij5JbmZvcm1hdGlvbiBhbmQgQ29tcHV0aW5nIFNjaWVuY2VzPC9zdWJqZWN0PgogICAgPHN1YmplY3Q+Q29tcHV0ZXIgdmlzaW9uPC9zdWJqZWN0PgogICAgPHN1YmplY3Q+QWN0aW9uIHJlY29nbml0aW9uPC9zdWJqZWN0PgogICAgPHN1YmplY3Q+M0QtQ05OPC9zdWJqZWN0PgogICAgPHN1YmplY3Q+QmFnTmV0PC9zdWJqZWN0PgogICAgPHN1YmplY3Q+VGVtcG9yYWwgcmVjZXB0aXZlIGZpZWxkPC9zdWJqZWN0PgogIDwvc3ViamVjdHM+CiAgPGNvbnRyaWJ1dG9ycz4KICAgIDxjb250cmlidXRvciBjb250cmlidXRvclR5cGU9Ik90aGVyIj4KICAgICAgPGNvbnRyaWJ1dG9yTmFtZSBuYW1lVHlwZT0iT3JnYW5pemF0aW9uYWwiPlRVIERlbGZ0LCBUVSBEZWxmdCwgRmFjdWx0eSBvZiBFbGVjdHJpY2FsIEVuZ2luZWVyaW5nLCBNYXRoZW1hdGljcyBhbmQgQ29tcHV0ZXIgU2NpZW5jZSwgRGVwYXJ0bWVudCBvZiBJbnRlbGxpZ2VudCBTeXN0ZW1zLCBDb21wdXRlciBWaXNpb24gTGFiPC9jb250cmlidXRvck5hbWU+CiAgICA8L2NvbnRyaWJ1dG9yPgogICAgPGNvbnRyaWJ1dG9yIGNvbnRyaWJ1dG9yVHlwZT0iT3RoZXIiPgogICAgICA8Y29udHJpYnV0b3JOYW1lIG5hbWVUeXBlPSJPcmdhbml6YXRpb25hbCI+VE5PLCBOZXRoZXJsYW5kcyBPcmdhbmlzYXRpb24gZm9yIEFwcGxpZWQgU2NpZW50aWZpYyBSZXNlYXJjaCwgSW50ZWxsaWdlbnQgSW1hZ2luZyBHcm91cDwvY29udHJpYnV0b3JOYW1lPgogICAgPC9jb250cmlidXRvcj4KICA8L2NvbnRyaWJ1dG9ycz4KICA8ZGF0ZXM+CiAgICA8ZGF0ZSBkYXRlVHlwZT0iSXNzdWVkIj4yMDI0LTA1LTI0PC9kYXRlPgogIDwvZGF0ZXM+CiAgPGxhbmd1YWdlPmVuPC9sYW5ndWFnZT4KICA8cmVsYXRlZElkZW50aWZpZXJzPgogICAgPHJlbGF0ZWRJZGVudGlmaWVyIHJlbGF0ZWRJZGVudGlmaWVyVHlwZT0iRE9JIiByZWxhdGlvblR5cGU9IklzU3VwcGxlbWVudFRvIj5odHRwczovL2RvaS5vcmcvMTAuMTEwOS9JQ0NWVzYwNzkzLjIwMjMuMDAwMjM8L3JlbGF0ZWRJZGVudGlmaWVyPgogIDwvcmVsYXRlZElkZW50aWZpZXJzPgogIDxmb3JtYXRzPgogICAgPGZvcm1hdD5HaXRIdWIgcmVwb3NpdG9yeSB3aXRoIHB5dGhvbiBjb2RlPC9mb3JtYXQ+CiAgPC9mb3JtYXRzPgogIDx2ZXJzaW9uPjE8L3ZlcnNpb24+CiAgPHJpZ2h0c0xpc3Q+CiAgICA8cmlnaHRzIHJpZ2h0c1VSST0iaHR0cHM6Ly9jcmVhdGl2ZWNvbW1vbnMub3JnL3B1YmxpY2RvbWFpbi96ZXJvLzEuMC8iPkNDMDwvcmlnaHRzPgogIDwvcmlnaHRzTGlzdD4KICA8ZGVzY3JpcHRpb25zPgogICAgPGRlc2NyaXB0aW9uIGRlc2NyaXB0aW9uVHlwZT0iQWJzdHJhY3QiPiZsdDtwJmd0OyZsdDtzcGFuIHN0eWxlPSJjb2xvcjogcmdiKDM0LCAzNCwgMzQpOyImZ3Q7UHJldmlvdXMgd29yayBvbiBsb25nLXRlcm0gdmlkZW8gYWN0aW9uIHJlY29nbml0aW9uIHJlbGllcyBvbiBkZWVwIDNELWNvbnZvbHV0aW9uYWwgbW9kZWxzIHRoYXQgaGF2ZSBhIGxhcmdlIHRlbXBvcmFsIHJlY2VwdGl2ZSBmaWVsZCAoUkYpLiBXZSBhcmd1ZSB0aGF0IHRoZXNlIG1vZGVscyBhcmUgbm90IGFsd2F5cyB0aGUgYmVzdCBjaG9pY2UgZm9yIHRlbXBvcmFsIG1vZGVsaW5nIGluIHZpZGVvcy4gQSBsYXJnZSB0ZW1wb3JhbCByZWNlcHRpdmUgZmllbGQgYWxsb3dzIHRoZSBtb2RlbCB0byBlbmNvZGUgdGhlIGV4YWN0IHN1Yi1hY3Rpb24gb3JkZXIgb2YgYSB2aWRlbywgd2hpY2ggY2F1c2VzIGEgcGVyZm9ybWFuY2UgZGVjcmVhc2Ugd2hlbiB0ZXN0aW5nIHZpZGVvcyBoYXZlIGEgZGlmZmVyZW50IHN1Yi1hY3Rpb24gb3JkZXIuIEluIHRoaXMgd29yaywgd2UgaW52ZXN0aWdhdGUgd2hldGhlciB3ZSBjYW4gaW1wcm92ZSB0aGUgbW9kZWwgcm9idXN0bmVzcyB0byB0aGUgc3ViLWFjdGlvbiBvcmRlciBieSBzaHJpbmtpbmcgdGhlIHRlbXBvcmFsIHJlY2VwdGl2ZSBmaWVsZCBvZiBhY3Rpb24gcmVjb2duaXRpb24gbW9kZWxzLiBGb3IgdGhpcywgd2UgZGVzaWduIFZpZGVvIEJhZ05ldCwgYSB2YXJpYW50IG9mIHRoZSAzRCBSZXNOZXQtNTAgbW9kZWwgd2l0aCB0aGUgdGVtcG9yYWwgcmVjZXB0aXZlIGZpZWxkIHNpemUgbGltaXRlZCB0byAxLCA5LCAxNyBvciAzMyBmcmFtZXMuIFdlIGFuYWx5emUgVmlkZW8gQmFnLU5ldCBvbiBzeW50aGV0aWMgYW5kIHJlYWwtd29ybGQgdmlkZW8gZGF0YXNldHMgYW5kIGV4cGVyaW1lbnRhbGx5IGNvbXBhcmUgbW9kZWxzIHdpdGggdmFyeWluZyB0ZW1wb3JhbCByZWNlcHRpdmUgZmllbGRzLiBXZSBmaW5kIHRoYXQgc2hvcnQgcmVjZXB0aXZlIGZpZWxkcyBhcmUgcm9idXN0IHRvIHN1Yi1hY3Rpb24gb3JkZXIgY2hhbmdlcywgd2hpbGUgbGFyZ2VyIHRlbXBvcmFsIHJlY2VwdGl2ZSBmaWVsZHMgYXJlIHNlbnNpdGl2ZSB0byB0aGUgc3ViLWFjdGlvbiBvcmRlci4gJmx0Oy9zcGFuJmd0OyZsdDtzcGFuIHN0eWxlPSJiYWNrZ3JvdW5kLWNvbG9yOiByZ2IoMjU1LCAyNTAsIDIzNCk7IiZndDtJbiB0aGlzIHJlcG9zaXRvcnksIHdlIHByb3ZpZGUgb3VyIGMmbHQ7L3NwYW4mZ3Q7Jmx0O3NwYW4gc3R5bGU9ImNvbG9yOiByZ2IoMzEsIDM1LCA0MCk7IiZndDtvZGUsIGluY2x1ZGluZyB0aGUgaW1wbGVtZW50YXRpb24gb2YgVmlkZW8gQmFnLU5ldC4mbHQ7L3NwYW4mZ3Q7Jmx0Oy9wJmd0OzwvZGVzY3JpcHRpb24+CiAgPC9kZXNjcmlwdGlvbnM+CjwvcmVzb3VyY2U+","url":"https://data.4tu.nl/datasets/dc5e2fb8-6005-40cd-9afa-ff03c57d0a23","contentUrl":null,"metadataVersion":0,"schemaVersion":"http://datacite.org/schema/kernel-4","source":"api","isActive":true,"state":"findable","reason":null,"viewCount":0,"viewsOverTime":[],"downloadCount":0,"downloadsOverTime":[],"referenceCount":0,"citationCount":1,"citationsOverTime":[{"year":"2024","total":1}],"partCount":0,"partOfCount":0,"versionCount":0,"versionOfCount":0,"created":"2024-05-24T10:53:17.000Z","registered":"2024-05-24T10:53:18.000Z","published":"2024","updated":"2024-05-24T18:22:53.000Z"},"relationships":{"client":{"data":{"id":"delft.data4tu","type":"clients"}},"provider":{"data":{"id":"scpr","type":"providers"}},"media":{"data":{"id":"10.4121/dc5e2fb8-6005-40cd-9afa-ff03c57d0a23","type":"media"}},"references":{"data":[]},"citations":{"data":[{"id":"10.1109/iccvw60793.2023.00023","type":"dois"}]},"parts":{"data":[]},"partOf":{"data":[]},"versions":{"data":[]},"versionOf":{"data":[]}}}}