Amateur: Augmented Reality Based Vehicle Navigation System Chu Cao 1 - - PowerPoint PPT Presentation

▶
amateur augmented reality based vehicle navigation system
SMART_READER_LITE
LIVE PREVIEW

Amateur: Augmented Reality Based Vehicle Navigation System Chu Cao 1 - - PowerPoint PPT Presentation

Mar 04, 2023 420 likes •859 views

Amateur: Augmented Reality Based Vehicle Navigation System Chu Cao 1 , Zhenjiang Li 2 , Pengfei Zhou 1 , Mo Li 1 Nanyang Technological University 1 , Singapore City University of Hong Kong 2 , Hong Kong, China 12 September 2019 UbiComp19,

slide-1
SLIDE 1

Amateur: Augmented Reality Based Vehicle Navigation System

Chu Cao 1, Zhenjiang Li 2, Pengfei Zhou 1, Mo Li 1

Nanyang Technological University 1, Singapore City University of Hong Kong 2, Hong Kong, China

UbiComp’19, London, UK 12 September 2019

slide-2
SLIDE 2

Vehicular navigation system Mobile navigation service

Transportation System

2

slide-3
SLIDE 3

Transportation System

Problem: display digital map Gap: real v.s. virtual

3

slide-4
SLIDE 4

AR-based Navigation Service

✦ Display front-view road condition ✦ Instructions on live world ✦ Comparable navigation ✦ Easy to deploy 4

slide-5
SLIDE 5

AR-based Navigation Service

❖ Challenges

✦ Host-lane identifying ✦ Annotations placement

Lane-level localisation accuracy Depth information in video

Determine the correct instruction at proper position on screen.

5

slide-6
SLIDE 6

AR-based Navigation Service

❖ Related work

CarLoc: Precisely Tracking Automobile Position [SenSys’15] Real time Detection of Lane Markers in Urban Streets [IVS’08]

✦

20 built-in sensors

✦

Dead-reckoning

✦

Complicated image processing

✦

Too heavy to be affordable

6

slide-7
SLIDE 7

AR-based Navigation Service

❖ Related work

Tesla autopilot 2.0 Google automobile

✦

Maintain on hostlane

✦

Rich sensor embedded

7

slide-8
SLIDE 8

AR-based Navigation Service

❖ Related work

Towards Unified Depth and Semantic Prediction from a Single Image [CVPR’15]

✦

Three complicated neural networks

✦

Large volume of training data

8

slide-9
SLIDE 9

Instructional Sign Placement Lane Identification Intersection Inference

Intersection Pin-hole Model Traffic Light Detection Subtracter Peak Detection Host Lane Particle Filter Frame Slice Instruction Type Instruction Position

AR-based Navigation Service

❖ System architecture

System

Navigation GPS User Input Lane-Road Information Origin and Destination

ce

Result AR GPS & Motion sensor Live video

Frames

9

slide-10
SLIDE 10

AR-based Navigation Service

❖ System architecture

Navigation

Instructional Sign Placement

GPS User Input

Lane Identification Intersection Inference

Result AR GPS & Motion sensor Live video Intersection Pin-hole Model Traffic Light Detection Subtracter Peak Detection Host Lane Particle Filter Frame Slice Instruction Type Instruction Position

Frames

Lane-Road Information Origin and Destination

10

slide-11
SLIDE 11

AR-based Navigation Service

❖ System architecture — lane identification

Lane detection task. Based on pure videos. Avoid collisions for automobiles. Lane identification task. Based not only on videos… IMU sensors on mobile phone & Extra lane number information Assistant for drivers.

11

slide-12
SLIDE 12

AR-based Navigation Service

❖ System architecture — lane identification

One frame in video Image slicing of 60 frames

12

slide-13
SLIDE 13

AR-based Navigation Service

❖ System architecture — lane identification

200 400 600 800 1000 1200 1400 1600 1800 50 100 150 200

Grey value Pixel indicator

h p w

1 2 3 4 5 6 7

h ≥ 1 2max{gi}n

i=1

<latexit sha1_base64="9RDv2MYFcOXvS2/hoKTrS5Q6S04=">ACDHicdVDLSgMxFM34rPVdekmWARBKDNtoXYhFNy4rGAf0Kklk2ba0MzDJCOWMB+gC5f+gB/gxoUibv0Ad36A/2FmWkFDwQO5zLzT1OyKiQpvluzMzOzS8sZpayura+u5jc2mCKOSQMHLOBtBwnCqE8akpG2iEnyHMYaTmjo8RvXRAuaOCfynFIuh4a+NSlGEkt9XL5oT0g59B2OcLKilUx9tClrQY9asc9RQ+t+CxJmYVqCjghlfKUVC1oFcwU+dr+XfP247pU7+Xe7H6AI4/4EjMkRMcyQ9lViEuKGYmzdiRIiPAIDUhHUx95RHRVekwMd7XSh27A9fMlTNXvEwp5Qow9Ryc9JIfit5eIf3mdSLoHXUX9MJLEx5NFbsSgDGDSDOxTrBkY0Q5lT/FeIh0r1I3V9Wl/B1KfyfNIsFq1Qonug2ymCDNgGO2APWKACauAY1EDYHAF7sEjeDJujAfj2XiZRGeM6cwW+AHj9ROtdJ+1</latexit>

Brightness

p = h − min{gi}

gk+ w

2

gk− w

2

<latexit sha1_base64="Rkf7FqIuQSgMdOn9OZY+TVWMXbI=">ACG3icdZDLSgMxFIYz9VbrerSTVBEQVpmxkLtQi40GUF2yqdWjJpg3NZIYko5Rh3sONj6IbF4q4ElwIfRjTi2BFfwj8fOcTs7vhoxKZqfRmpmdm5+Ib2YWVpeWV3Lrm/UZBAJTKo4YIG4dJEkjHJSVQxchkKgnyXkbrbOxnW6zdESBrwC9UPSdNHU49ipHSqJW1w+NuzqfciTst6iTXce6d+B4AuH4NontJGkNSW6KZHfMfGkODbFwsSULGjlzZF2yvtXp3uDh0GlX132gGOfMIVZkjKhmWGqhkjoShmJMk4kSQhwj3UIQ1tOfKJbMaj2xK4q0kbeoHQjys4oj8nYuRL2fd3ekj1ZW/a0P4V60RKe+oGVMeRopwPF7kRQyqA6Dgm0qCFasrw3Cguq/QtxFOgal48zoEL4vhf+bmp23DvP2uU6jAMZKgy2wDfaBYqgDM5ABVQBnfgETyDF+PeDJejbdxa8qYzGyCKRkfX4oprQ=</latexit>

1 3p ≤ p − G10th

<latexit sha1_base64="SZ0J2yGuG58XxklAN9wP4PnaU=">ACB3icdVDLSsNAFJ34rPUVdSnIYBEsSRtoXZXcKHLCvYBbQiT6aQdOnk4MxFKyM6NfoA/4M6NC0Xc+gvu/AD/w2lSQUPXDicy/3uOEjApGO/azOzc/MJibim/vLK6tq5vbLZEHFMmjhgAe84SBGfdKUVDLSCTlBnsNI2xkdT/z2JeGCBv65HIfE8tDApy7FSCrJ1nd6Lkc4NpO4nISwx8gFjMPDEzs2DTlMElsvGMVaCpiRamVKaiY0i0aKQv3grnX7cVNu2Ppbrx/gyCO+xAwJ0TWNUFox4pJiRpJ8LxIkRHiEBqSrqI8Iqw4/SOBe0rpQzfgqnwJU/X7RIw8Icaeozo9JIfitzcR/K6kXSPrJj6YSJj7NFbsSgDOAkFNinGDJxogzKm6FeIhUsFIFV1ehfD1KfyftEpFs1wsnak0KiBDmyDXbAPTFAFdXAKGqAJMLgC9+ARPGnX2oP2rL1krTPadGYL/ID2+gnkFp0C</latexit>

Sharpness

13

slide-14
SLIDE 14

Lane Number Lane Marker Number Lane Direction [-

AR-based Navigation Service

❖ System architecture — lane identification

A road segment with 4 lanes Templates of a 4-lane road

[474] [790] [1106] [474] [790] [-474]

X1

[474] [790] [-474]

X2

[474] [-474] [-790]

X3

[-474] [-790] [-1106]

X4

14

slide-15
SLIDE 15

AR-based Navigation Service

❖ System architecture — lane identification

Ideally detect the peaks

Matching Xobs X3

[474] [-474] [-790] [474] [-474] [-790]

1. Blockage of frontal vehicles

  • 2. Reflection of lights
  • 3. Bad condition of lane markers

15

slide-16
SLIDE 16

AR-based Navigation Service

❖ System architecture — lane identification

✦ Particle filter design

[474] [-474] [-790] [790]

Xobs

[474] [790] [1106] [474] [790] [-474]

X1

[474] [790] [-474]

X2

[474] [-474] [-790]

X3

[-474] [-790] [-1106]

X4

initialisation

Dynamic time wrapping Weight updating

16

slide-17
SLIDE 17

AR-based Navigation Service

❖ System architecture — lane identification

wp

b = e−db

<latexit sha1_base64="eXOYNe48FvqEsizy0aM4VtsE/i0=">AB9XicdVDLSsNAFJ3UV62vqhvBzWAR3BiSWqhFhIblxXsA9o0TCaTdujkwczEUkL3foIbF4q4danf4c6/cZpUNEDFw7n3Mu9zgRo0IaxoeW1hcWl7JrxbW1jc2t4rbOy0RxhyTJg5ZyDsOEoTRgDQlYx0Ik6Q7zDSdkYXM79Q7igYXAtJxGxfDQIqEcxkrqj2nH52TfnLs2s7ULpYMvZYCZqRamZOaCU3dSFGq791ar6Ozt4ZdfO+5IY59EkjMkBd04iklSAuKWZkWujFgkQIj9CAdBUNkE+ElaRXT+GhUlzohVxVIGqfp9IkC/ExHdUp4/kUPz2ZuJfXjeW3qmV0CKJQlwtsiLGZQhnEUAXcoJlmyiCMKcqlshHiKOsFRBFVQIX5/C/0mrJsnevlKpVEBGfJgHxyAI2CKqiDS9ATYAB3fgATxqY+1e9Kes9acNp/ZBT+gvXwCiCWLQ=</latexit>

[474] [-474] [-790] [790]

Matching Xobs X3

[474] [-474] [-790]

✦ Particle filter design

Dynamic time wrapping under Euclidean distance

17

slide-18
SLIDE 18

AR-based Navigation Service

❖ System architecture — lane identification

Movement: lane switching

✦ Particle filter design

Resampling based on importance

[474] [-474] [-790] [790]

Xobs

18

slide-19
SLIDE 19

AR-based Navigation Service

❖ System architecture — lane identification

✦ Particle filter design

Resampling based on importance

19

slide-20
SLIDE 20

AR-based Navigation Service

❖ System architecture — lane identification

✦ Particle filter design

Resampling based on importance Lane marker of host-lane has a traversing phenomenon during lane switching.

20

slide-21
SLIDE 21

AR-based Navigation Service

❖ System architecture — lane identification P(Ls|O) = P(O|Ls) · P(Ls) P(O)

<latexit sha1_base64="OilGbpP5ivGFm41K08aA2r/hG9I=">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</latexit>

P(Ls|O) = P(O|Ls) · β

<latexit sha1_base64="F+5Sx7dI6tf6U3W8FKAEb7P86qI=">ACQHicdVBLSwMxGMz6tr6qHr0Ei9Bel0t1B6EghcPghWsFbqlZNsG5p9kHwrlO3+NC/+BG+evXhQxKsn03YVlToQmMx8Q76MGwmuwLIejbn5hcWl5ZXV3Nr6xuZWfnvnWoWxpKxBQxHKG5coJnjAGsBsJtIMuK7gjXdwenYb94yqXgYXMEwYm2f9ALucUpAS5180/EJ9CkRST0tfvPztKNG37eLtHQyc+wiHf2KlBzaDcFxGZBOvmCZ1QnwlFTKGana2DatCQoQ72Tf3C6IY19FgAVRKmWbUXQTogETgVLc06sWETogPRYS9OA+Ey1k0kBKT7QShd7odQnADxRfyYS4is19F09Od5X/fXG4iyvFYN3E54EMXAjp9yIsFhCP28RdLhkFMdSEUMn1rpj2iSQUdOc5XcLXT/H/5PrQtI/Mw8tyoWZmdaygPbSPishGFVRDZ6iOGoiO/SEXtCrcW8G2/G+3R0zsgyu+gXjI9P65iy9w=</latexit>

P(O|Ls) = e−d0

s

<latexit sha1_base64="2WczuQIuT4KbxgclCcSKZ6L9I+0=">ACHXicdVDLSgMxFM3UV62v+ti5CRaxLhxmaqF2IRTcuBCsYGuhHYdMJtOGZh4kGaGM8yNu/BU3LhRx4Ub8G9OHRUPBE7OuZd73EiRoU0jA8tMzM7N7+QXcwtLa+sruXN5oijDkmDRykLcJAijAWlIKhlpRZwg32HkyumfDP2rG8IFDYNLOYiI5aNuQD2KkVSnS93fCR7GLGknhan/DyFt3D6O0tsQ+PIblODlxb7KV2vmDo1RHgmFTKE1I1oakbIxTABHU7/9ZxQxz7JCYISHaphFJK0FcUsxImuvEgkQI91GXtBUNkE+ElYyuS+GuUlzohVy9QMKR+r0jQb4QA9RlcONxW9vKP7ltWPpHVkJDaJYkgCPB3kxgzKEw6igSznBkg0UQZhTtSvEPcQRlirQnArh61L4P2mWdPNQL12UC7WtSRxZsA12QBGYoAJq4BTUQNgcAcewBN41u61R+1Fex2XZrRJzyb4Ae39E8DzojI=</latexit>

✦ Particle filter design

Resampling based on importance

21

slide-22
SLIDE 22

AR-based Navigation Service

❖ System architecture — lane identification

✦ Particle filter design

Resampling based on importance

ˆ wl = X

s+i=l

P(Ls|O) · wi = X

s+i=l

β · P(O|Ls) · wi

<latexit sha1_base64="F35vbOycUXeIgJTW5UQYFwL5VQ=">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</latexit>

22

slide-23
SLIDE 23

Navigation

Instructional Sign Placement

GPS User Input

Lane Identification Intersection Inference

Result AR GPS & Motion sensor Live video Intersection Pin-hole Model Traffic Light Detection Subtracter Peak Detection Host Lane Particle Filter Frame Slice Instruction Type Instruction Position

Frames

Lane-Road Information Origin and Destination

AR-based Navigation Service

❖ System architecture

Instructional S Lane Identification

Peak Detection Host Lane Particle Filter Frame Slice Instruction Type

23

slide-24
SLIDE 24

Navigation

Instructional Sign Placement

GPS User Input

Lane Identification Intersection Inference

Result AR GPS & Motion sensor Live video Intersection Pin-hole Model Traffic Light Detection Subtracter Peak Detection Host Lane Particle Filter Frame Slice Instruction Type Instruction Position

Frames

Lane-Road Information Origin and Destination

AR-based Navigation Service

❖ System architecture

Instructional S Lane Identification

Peak Detection Host Lane Particle Filter Frame Slice Instruction Type

✓

24

slide-25
SLIDE 25

Navigation

Instructional Sign Placement

GPS User Input

Lane Identification Intersection Inference

Result AR GPS & Motion sensor Live video Intersection Pin-hole Model Traffic Light Detection Subtracter Peak Detection Host Lane Particle Filter Frame Slice Instruction Type Instruction Position

Frames

Lane-Road Information Origin and Destination

AR-based Navigation Service

❖ System architecture

Instructional S Lane Identification

Peak Detection Host Lane Particle Filter Frame Slice Instruction Type

✓

25

slide-26
SLIDE 26

Navigation

Instructional Sign Placement

GPS User Input

Lane Identification Intersection Inference

Result AR GPS & Motion sensor Live video Intersection Pin-hole Model Traffic Light Detection Subtracter Peak Detection Host Lane Particle Filter Frame Slice Instruction Type Instruction Position

Frames

Lane-Road Information Origin and Destination

AR-based Navigation Service

❖ System architecture

al Sign Placement Intersection Inference

Intersection Pin-hole Model Traffic Light Detection Subtracter Instruction Position

26

Please refer to our paper for more details.

slide-27
SLIDE 27

Navigation

Instructional Sign Placement

GPS User Input

Lane Identification Intersection Inference

Result AR GPS & Motion sensor Live video Intersection Pin-hole Model Traffic Light Detection Subtracter Peak Detection Host Lane Particle Filter Frame Slice Instruction Type Instruction Position

Frames

Lane-Road Information Origin and Destination

AR-based Navigation Service

❖ System architecture

Instructional Sign Placement Lane Identification Intersection Inference

Intersection Pin-hole Model Traffic Light Detection Subtracter Peak Detection Host Lane Particle Filter Frame Slice Instruction Type Instruction Position

27

slide-28
SLIDE 28

AR-based Navigation Service

❖ Examples

1 2 2 2

Cloudy Daytime Nightfall

28

slide-29
SLIDE 29

AR-based Navigation Service

❖ Demo

29

slide-30
SLIDE 30

AR-based Navigation Service

❖ Evaluation

30

slide-31
SLIDE 31

AR-based Navigation Service

❖ Evaluation

✦ Running on rental taxis ✦ Levenshtein distance ✦ Ground truth: manually checking the video ✦ Implemented on Nexus 5X

Around 300 km travel distance Vgen = {LC2, L, RC1, R}, Vgt = {LC1, L, RC1, R}

31

slide-32
SLIDE 32

AR-based Navigation Service

❖ Evaluation

acc = Levenshtein |Vgt|

<latexit sha1_base64="PHUkX7ZaVIUey43QnpEritsaU=">ACXicdVA9SwNBEN2LXzF+RS1tFoNgFe5iQC2EgI2FhYJhCSEvc1csmRv79idE8J5rY1/xcZCEVv/gZ3/xs2HoKIPBh7vzTAz4+lMOi6H05ubn5hcSm/XFhZXVvfKG5uNUyUaA51HslIX/vMgBQK6ihQwnWsgYW+hKY/PB37zRvQRkTqCkcxdELWVyIQnKGVukXKOD9pB5rx9BxuQJkBglBZetvopn3MbrNuseSWjyegU3JYnZFj3pld4ISmeGiW3xv9yKehKCQS2ZMy3Nj7KRMo+ASskI7MRAzPmR9aFmqWAimk04+yeieVXo0iLQthXSifp9IWjMKPRtZ8hwYH57Y/Evr5VgcNRJhYoTBMWni4JEUozoOBbaExo4ypEljGthb6V8wGwsaMr2BC+PqX/k0al7B2UK5fVUq06iyNPdsgu2SceOSQ1ckYuSJ1wckceyBN5du6dR+fFeZ25pzZzDb5AeftE+0Um8U=</latexit>

43 74 101 67 86 # of instructions

20 40 60 80 100 A B C D E Morning Peak Evening Peak Off-peak

Accuracy (%) Route

32

slide-33
SLIDE 33

AR-based Navigation Service

❖ Evaluation

33

slide-34
SLIDE 34

AR-based Navigation Service

❖ Evaluation

✦ Around 27m error (52m) ✦ 105 pixels offset on screen (127 pixels)

CDF 0.2 0.4 0.6 0.8 1 Turning Arrow Placement Offset on Screen (pixels) 20 40 60 80 100 120 140

(90, 0.9) (109, 0.9) Our system GPS readings

34

slide-35
SLIDE 35

Thank you very much.

Q & A

slide-36
SLIDE 36

Backup slides.

slide-37
SLIDE 37

AR-based Navigation Service

❖ System architecture — intersection inference

1. Flickering feature of LED bulbs 2. Rolling shutter effect on CMOS

RGB to Gray Subtracter Binarization Color filtering y of lights

Frames Frames Frames

37

slide-38
SLIDE 38

AR-based Navigation Service

❖ System architecture — intersection inference

ia = h h − H it

<latexit sha1_base64="5f7cJ9qrJCfMc3dVDs2dBiKBF48=">AB/XicdZBLS8NAFIUn9VXrKz52bgaL4MaQ1ELtQi46bKCfUAbwmQ6aYZOJmFmItRQ/CtuXCji1v/hzn/jtI1QRQ8MHM65l7l8fsKoVLb9aRWVtfWN4qbpa3tnd09c/+gI+NUYNLGMYtFz0eSMpJW1HFSC8RBEU+I1/fD3ru3dESBrzWzVJiBuhEacBxUjpyDOPqIeuBoFAOAunWXjenFJPeWbZsey5oG3VtWrV3NQd+F2VQa6WZ34MhjFOI8IVZkjKvmMnys2QUBQzMi0NUkShMdoRPrachQR6Wbz6fwVCdDGMRCP67gPF3eyFAk5STy9WSEVCh/d7Pwr6fquDSzShPUkU4XnwUpAyqGM5QwCEVBCs20QZhQfWtEIdIo1AaWGkZwv+mU7GcC6tyUy03qjmOIjgGJ+AMOKAGqAJWqANMLgHj+AZvBgPxpPxarwtRgtGvnMIfsh4/wIKapW</latexit>

The position of traffic light on the screen is known.

Image Plane Traffic Light D H h f it ia Focal Point A Road Surface

38

slide-39
SLIDE 39

AR-based Navigation Service

❖ Evaluation

✦ Until converging ✦ Accurately identifying in 5s ✦ Force output in 2s ✦ More observations ->

better performance

25 50 75 100 2 3 4 5 Accuracy Time

Accuracy (%) Latency (s)

2 1.5 1 0.5

Number of lanes Number of lanes

5 4 3 2

Latency (s)

1 2 3 4 5

Mean Standard deviation

39

slide-40
SLIDE 40

AR-based Navigation Service

❖ Evaluation

Likert scale rating questions

✦

Ease of use (Q1 & Q5)

✦

Perceived distraction (Q2 & Q6)

✦

Navigational experience (Q4 & Q8)

✦

User-friendliness (Q3 & Q7)

No. Rating Question Statement Q1 It was easy to navigate using this navigation service. Q2 I need to pay extra attention on this navigation service when driving. Q3 This navigation service provided user-friendly guidance . Q4 This navigation service was useful in helping me navigate properly. Q5 It was easy for me to learn how to use this navigation service. Q6 I paid most of my attention on driving using this navigation service. Q7 The guidance was user-friendly to interact with. Q8 This navigation service provided me with effective guidance.

40

slide-41
SLIDE 41

AR-based Navigation Service

❖ Evaluation

✦

With-in subject user study

✦

Wilcoxon Signed Rank Tests

✦

Easier to use

✦

Less distracted

✦

More user-friendly

No. Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 p-value 0.0016 <0.0001 <0.0001 0.0003 <0.0001 <0.0001 <0.0001 <0.0001

Scale Rating

1 2 3 4 5 6 7 8

Question Indicator

A 1 G 1 A 2 G 2 A 3 G 3 A 4 G 4 A 5 G 5 A 6 G 6 A 7 G 7 A 8 G 8

41

slide-42
SLIDE 42

Route A B C D E Length (km) 12.7 17.4 36.8 12.5 16.8 Length of Expressway (km) 8.73 15.35 36 4.46 14.28 Length of Highway (km) 3.97 2.05 0.8 8.04 2.52 Number of Traffic Lights 15 19 5 18 17 Average Velocity (km/h) 49.7 51.2 56.1 45.4 53.7

AR-based Navigation Service

❖ Evaluation — details of routes

42

slide-43
SLIDE 43

Property Description of Group A Description of Group B Age (year) 32 - 61 (mean: 40.3) 31 - 57 (mean: 42.8) Driving Experience (year) 2 - 27 (mean: 15.8) 1 - 30 (mean: 17.6) Gender 21 Male (84%), 4 Female (16%) 19 Male (76%), 6 Female (24%)

AR-based Navigation Service

❖ Evaluation — details of drivers

43