Person Re-Identification Yiheng Liu Outli line Background - - PowerPoint PPT Presentation

person re identification
SMART_READER_LITE
LIVE PREVIEW

Person Re-Identification Yiheng Liu Outli line Background - - PowerPoint PPT Presentation

Person Re-Identification Yiheng Liu Outli line Background Image-Based Person Re-Identification Partial Person Re-identification Video-Based Person Re-Identification Our Methods Outli line Background Image-Based


slide-1
SLIDE 1

Person Re-Identification

Yiheng Liu

slide-2
SLIDE 2

Outli line

  • Background
  • Image-Based Person Re-Identification
  • Partial Person Re-identification
  • Video-Based Person Re-Identification
  • Our Methods
slide-3
SLIDE 3

Outli line

  • Background
  • Image-Based Person Re-Identification
  • Partial Person Re-identification
  • Video-Based Person Re-Identification
  • Our Methods
slide-4
SLIDE 4

Person Re-id identif ification

  • Person re-identification aims to match persons across

non-overlapping surveillance camera views.

Images:https://www.tugraz.at/institute/icg/research/team-bischof/lrs/downloads/prid11/

数据 检测 跟踪 匹配

slide-5
SLIDE 5

Datasets and Protocols ls

Dataset ID Images Image Market1501 1501 32217 DukeMTMC-reID 1812 36441 CUHK03 1467 13164 MSMT17 4101 126441 Video iLIDS-VID 300 42495 PRID2011 934 24541 MARS 1261 1191003

  • Cumulated Matching Characteristics (CMC):Rank-1, Rank-5
  • mean Average Precision (mAP)
slide-6
SLIDE 6

Outli line

  • Background
  • Image-Based Person Re-Identification
  • Partial Person Re-identification
  • Video-Based Person Re-Identification
  • Our Methods
slide-7
SLIDE 7

Learning Dis iscrim iminative Features wit ith Mult ltiple Granularitie ies for Person Re-Id Identification

slide-8
SLIDE 8
slide-9
SLIDE 9

Beyond Part Models ls: Person Retrie ieval l wit ith Refin ined Part Poolin ing (and A Strong Convolutional l Baseline)

slide-10
SLIDE 10

Refin ined Part Poolin ing

slide-11
SLIDE 11
slide-12
SLIDE 12

Horiz izontal l Pyramid Matchin ing for Person Re-id identific ication

slide-13
SLIDE 13
slide-14
SLIDE 14

Learning In Incremental Trip iple let Margin for Person Re-id identific ication

slide-15
SLIDE 15
slide-16
SLIDE 16

Outli line

  • Background
  • Image-Based Person Re-Identification
  • Partial Person Re-identification
  • Video-Based Person Re-Identification
  • Our Methods
slide-17
SLIDE 17

Perceive Where to Focus: Learning Vis isibilit ity-aware Part-le level Features for Partial l Person Re-id identif ification

slide-18
SLIDE 18

Region feature: Overall distance:

slide-19
SLIDE 19
  • The training of region locator
  • Cross-entropy loss
  • Triplet loss
slide-20
SLIDE 20
slide-21
SLIDE 21
slide-22
SLIDE 22

Outli line

  • Background
  • Image-Based Person Re-Identification
  • Partial Person Re-identification
  • Video-Based Person Re-Identification
  • Our Methods
slide-23
SLIDE 23

STA: Spatial-Temporal l Attention for Large-Scale le Vid ideo-based Person Re-Id Identif ification

Inter-Frame Regularization

slide-24
SLIDE 24

Feature Fusion Strategy

slide-25
SLIDE 25
slide-26
SLIDE 26

THANK YOU FOR LISTENING