813a2ac4a038ecb3876467a749d630b7 sensors_a2014v14p4189.pdf c8ddc862201c8f2790f333510b31a7a2d3f4b954 sensors_a2014v14p4189.pdf 0138c8127fb40ac9a24cba342a2d53c922d7df4bc18580fce60c687e7829eae7 sensors_a2014v14p4189.pdf Title: A Survey on Model Based Approaches for 2D and 3D Visual Human Pose Recovery Subject: Human Pose Recovery has been studied in the field of Computer Vision for the last 40 years. Several approaches have been reported, and significant improvements have been obtained in both data representation and model design. However, the problem of Human Pose Recovery in uncontrolled environments is far from being solved. In this paper, we define a general taxonomy to group model based approaches for Human Pose Recovery, which is composed of five main modules: appearance, viewpoint, spatial relations, temporal consistence, and behavior. Subsequently, a methodological comparison is performed following the proposed taxonomy, evaluating current SoA approaches in the aforementioned five group categories. As a result of this comparison, we discuss the main advantages and drawbacks of the reviewed literature. Keywords: human pose recovery; human body modelling; behavior analysis; computer vision Author: Xavier Perez-Sala 1 ,*, Sergio Escalera 2, Cecilio Angulo 3, Jordi Gonz`alez 4 Creator: dvips(k) 5.992 Copyright 2012 Radical Eye Software Producer: GPL Ghostscript 9.05 CreationDate: Mon Mar 3 17:24:58 2014 ModDate: Tue Feb 23 16:16:29 2016 Tagged: no UserProperties: no Suspects: no Form: none JavaScript: no Pages: 22 Encrypted: no Page size: 595 x 842 pts (A4) Page rot: 0 File size: 863226 bytes Optimized: yes PDF version: 1.4 name type encoding emb sub uni object ID ------------------------------------ ----------------- ---------------- --- --- --- --------- AXWRVB+NimbusRomNo9L-Medi Type 1C Custom yes yes no 129 0 YLXWQD+NimbusRomNo9L-Regu Type 1C Custom yes yes no 133 0 QARSKI+NimbusRomNo9L-MediItal Type 1C WinAnsi yes yes no 135 0 WTBOIB+CMBX8 Type 1C WinAnsi yes yes no 137 0 QEOWAT+CMMIB8 Type 1C Custom yes yes no 141 0 WTBOIB+CMR8 Type 1C WinAnsi yes yes no 147 0 IDLVRG+NimbusRomNo9L-ReguItal Type 1C WinAnsi yes yes no 143 0 SDEKAO+CMSY10 Type 1C Custom yes yes yes 91 0 CXGUNB+CMMI12 Type 1C WinAnsi yes yes no 101 0 GBYOJQ+CMMI8 Type 1C WinAnsi yes yes no 104 0 Jhove (Rel. 1.6, 2011-01-04) Date: 2016-02-26 10:53:31 CET RepresentationInformation: sensors_a2014v14p4189.pdf ReportingModule: PDF-hul, Rel. 1.8 (2009-05-22) LastModified: 2016-02-23 16:17:04 CET Size: 863226 Format: PDF Version: 1.4 Status: Well-Formed and valid SignatureMatches: PDF-hul MIMEtype: application/pdf Profile: Linearized PDF, ISO PDF/A-1, Level B PDFMetadata: Objects: 160 FreeObjects: 1 IncrementalUpdates: 1 DocumentCatalog: PageLayout: SinglePage PageMode: UseThumbs Info: Title: A Survey on Model Based Approaches for 2D and 3D Visual Human Pose Recovery Author: Xavier Perez-Sala 1 ,*, Sergio Escalera 2, Cecilio Angulo 3, Jordi Gonz`alez 4 Subject: Human Pose Recovery has been studied in the field of Computer Vision for the last 40 years. Several approaches have been reported, and significant improvements have been obtained in both data representation and model design. However, the problem of Human Pose Recovery in uncontrolled environments is far from being solved. In this paper, we define a general taxonomy to group model based approaches for Human Pose Recovery, which is composed of five main modules: appearance, viewpoint, spatial relations, temporal consistence, and behavior. Subsequently, a methodological comparison is performed following the proposed taxonomy, evaluating current SoA approaches in the aforementioned five group categories. As a result of this comparison, we discuss the main advantages and drawbacks of the reviewed literature. Keywords: human pose recovery; human body modelling; behavior analysis; computer vision Creator: dvips(k) 5.992 Copyright 2012 Radical Eye Software Producer: GPL Ghostscript 9.05 CreationDate: Mon Mar 03 10:24:58 CET 2014 ModDate: Tue Feb 23 16:16:29 CET 2016 ID: 0x2091eff2229bffba9882437ccec3ad7b, 0x590be0d05f152f4abd584a27cb1c1b46 Filters: FilterPipeline: FlateDecode FilterPipeline: DCTDecode Images: Image: NisoImageMetadata: CompressionScheme: JPEG ImageWidth: 667 ImageHeight: 123 ColorSpace: RGB BitsPerSample: 8 BitsPerSampleUnit: integer Image: NisoImageMetadata: CompressionScheme: JPEG ImageWidth: 739 ImageHeight: 180 ColorSpace: RGB BitsPerSample: 8 BitsPerSampleUnit: integer Image: NisoImageMetadata: CompressionScheme: JPEG ImageWidth: 767 ImageHeight: 218 ColorSpace: RGB BitsPerSample: 8 BitsPerSampleUnit: integer Image: NisoImageMetadata: CompressionScheme: JPEG ImageWidth: 781 ImageHeight: 418 ColorSpace: RGB BitsPerSample: 8 BitsPerSampleUnit: integer Image: NisoImageMetadata: CompressionScheme: JPEG ImageWidth: 828 ImageHeight: 224 ColorSpace: RGB BitsPerSample: 8 BitsPerSampleUnit: integer Image: NisoImageMetadata: CompressionScheme: JPEG ImageWidth: 795 ImageHeight: 266 ColorSpace: RGB BitsPerSample: 8 BitsPerSampleUnit: integer Image: NisoImageMetadata: CompressionScheme: JPEG ImageWidth: 2082 ImageHeight: 3504 ColorSpace: RGB BitsPerSample: 8 BitsPerSampleUnit: integer Fonts: Type1: Font: BaseFont: WTBOIB+CMBX8 FontSubset: true FirstChar: 49 LastChar: 52 FontDescriptor: FontName: WTBOIB+CMBX8 Flags: Nonsymbolic, AllCap FontBBox: 0, -14, 578, 659 FontFile3: true Encoding: WinAnsiEncoding Font: BaseFont: CXGUNB+CMMI12 FontSubset: true FirstChar: 80 LastChar: 80 FontDescriptor: FontName: CXGUNB+CMMI12 Flags: Nonsymbolic, AllCap FontBBox: 0, 0, 740, 683 FontFile3: true Encoding: WinAnsiEncoding Font: BaseFont: QEOWAT+CMMIB8 FontSubset: true FirstChar: 59 LastChar: 59 FontDescriptor: FontName: QEOWAT+CMMIB8 Flags: Symbolic, AllCap FontBBox: 0, -194, 264, 160 FontFile3: true EncodingDictionary: BaseEncoding: WinAnsiEncoding Differences: true Font: BaseFont: IDLVRG+NimbusRomNo9L-ReguItal FontSubset: true FirstChar: 40 LastChar: 122 FontDescriptor: FontName: IDLVRG+NimbusRomNo9L-ReguItal Flags: Serif, Nonsymbolic, Italic FontBBox: -147, -209, 873, 683 FontFile3: true Encoding: WinAnsiEncoding Font: BaseFont: AXWRVB+NimbusRomNo9L-Medi FontSubset: true FirstChar: 2 LastChar: 122 FontDescriptor: FontName: AXWRVB+NimbusRomNo9L-Medi Flags: Serif, Symbolic FontBBox: 0, -206, 921, 713 FontFile3: true EncodingDictionary: BaseEncoding: WinAnsiEncoding Differences: true Font: BaseFont: WTBOIB+CMR8 FontSubset: true FirstChar: 49 LastChar: 52 FontDescriptor: FontName: WTBOIB+CMR8 Flags: Nonsymbolic, AllCap FontBBox: 0, -21, 500, 675 FontFile3: true Encoding: WinAnsiEncoding Font: BaseFont: YLXWQD+NimbusRomNo9L-Regu FontSubset: true FirstChar: 2 LastChar: 180 FontDescriptor: FontName: YLXWQD+NimbusRomNo9L-Regu Flags: Serif, Symbolic FontBBox: -70, -218, 932, 688 FontFile3: true EncodingDictionary: BaseEncoding: WinAnsiEncoding Differences: true Font: BaseFont: GBYOJQ+CMMI8 FontSubset: true FirstChar: 105 LastChar: 105 FontDescriptor: FontName: GBYOJQ+CMMI8 Flags: Nonsymbolic, SmallCap FontBBox: 0, -10, 317, 662 FontFile3: true Encoding: WinAnsiEncoding Font: BaseFont: SDEKAO+CMSY10 FontSubset: true FirstChar: 13 LastChar: 24 FontDescriptor: FontName: SDEKAO+CMSY10 Flags: Symbolic FontBBox: 0, -216, 944, 716 FontFile3: true EncodingDictionary: BaseEncoding: WinAnsiEncoding Differences: true ToUnicode: true Font: BaseFont: QARSKI+NimbusRomNo9L-MediItal FontSubset: true FirstChar: 101 LastChar: 115 FontDescriptor: FontName: QARSKI+NimbusRomNo9L-MediItal Flags: Nonsymbolic, SmallCap FontBBox: -21, -13, 493, 462 FontFile3: true Encoding: WinAnsiEncoding XMP: GPL Ghostscript 9.05 human pose recovery; human body modelling; behavior analysis; computer vision 2016-02-23T16:16:29+01:00 2014-03-03T17:24:58+08:00 dvips(k) 5.992 Copyright 2012 Radical Eye Software 2016-02-23T16:16:29+01:00 uuid:30125a52-a511-11e3-0000-e9456075f466 uuid:5b57c0aa-dc39-40d3-8b01-8913ee6b5c44 application/pdf A Survey on Model Based Approaches for 2D and 3D Visual Human Pose Recovery Xavier Perez-Sala 1 ,*, Sergio Escalera 2, Cecilio Angulo 3, Jordi Gonz`alez 4 Human Pose Recovery has been studied in the field of Computer Vision for the last 40 years. Several approaches have been reported, and significant improvements have been obtained in both data representation and model design. However, the problem of Human Pose Recovery in uncontrolled environments is far from being solved. In this paper, we define a general taxonomy to group model based approaches for Human Pose Recovery, which is composed of five main modules: appearance, viewpoint, spatial relations, temporal consistence, and behavior. Subsequently, a methodological comparison is performed following the proposed taxonomy, evaluating current SoA approaches in the aforementioned five group categories. As a result of this comparison, we discuss the main advantages and drawbacks of the reviewed literature. human pose recovery human body modelling behavior analysis computer vision Pages: Page: Sequence: 1 Page: Sequence: 2 Page: Sequence: 3 Page: Sequence: 4 Page: Sequence: 5 Page: Sequence: 6 Page: Sequence: 7 Page: Sequence: 8 Page: Sequence: 9 Page: Sequence: 10 Page: Sequence: 11 Page: Sequence: 12 Page: Sequence: 13 Page: Sequence: 14 Rotate: 90 Page: Sequence: 15 Page: Sequence: 16 Page: Sequence: 17 Page: Sequence: 18 Page: Sequence: 19 Page: Sequence: 20 Page: Sequence: 21 Page: Sequence: 22 Checksum: 871535b6 Type: CRC32 Checksum: 813a2ac4a038ecb3876467a749d630b7 Type: MD5 Checksum: c8ddc862201c8f2790f333510b31a7a2d3f4b954 Type: SHA-1