Photographer anonymity Joint work with Peter Schaffer (Uni. of - - PowerPoint PPT Presentation

▶

Jan 21, 2023 37 likes •348 views

Photographer anonymity Joint work with Peter Schaffer (Uni. of Luxembourg) Djamila Aouada (Uni. of Luxembourg) Presented by: Shishir Nagaraja (IIIT Delhi) nagaraja@iiitd.ac.in WPES 2011 The perils of being on camera Person X is falsely on

SLIDE 1

Photographer anonymity

Joint work with

Peter Schaffer (Uni. of Luxembourg) Djamila Aouada (Uni. of Luxembourg) Presented by: Shishir Nagaraja (IIIT Delhi) nagaraja@iiitd.ac.in

WPES 2011

SLIDE 2

The perils of being on camera

Person X is falsely on medical-disability leave but is visiting the UK Person Y happens to be visiting UK and accidently takes pictures of X and posts it on the Internet

Defenses: Pixellation, blurring, ...

SLIDE 3

A different privacy problem: photographer anonymity

Citizen journalist Secret meeting

1. Photograph
2. Blog article

SLIDE 4

Photographer anonymity problem

Citizen journalist with hidden camera Secret meeting

1. Photograph
2. Blog article

SLIDE 5

Naive defense: hidden cameras

SLIDE 6

Hidden photographer

Anonymity set with hidden camera

SLIDE 7

Citizen journalist with hidden camera Secret meeting

1. Photograph
2. Post image
n the Internet
3. Image analysis to

determine hidden camera location

4. Consult

CCTV to deanonymise photographer

Camera-location De-anonymisation attack

SLIDE 8

Our approach towards a defense technique: View Synthesis

Source: L. Fei-fei, Stanford Vision Lab

SLIDE 9

Hidden photographer1 Hidden photographer2

Anonymity set with view-synthesis

SLIDE 10

Citizen journalist with hidden camera Secret meeting

1. Photograph
2. Post image
n the Internet
3. Image analysis to

determine hidden camera location

4. Consult

CCTV to deanonymise photographer

Camera-location De-anonymisation attack

?

SLIDE 11

Problem definition

Input:
Two images taken from locations L and R respectively
Two projection matrices specifying image viewpoints
Output projection matrix specifying synthetic viewpoint
Output: synthetic image
Security properties: minimal information about L or R

should be leaked

Threat model: external passive adversary

SLIDE 12

Privacy enhancing technology: view synthesis

SLIDE 13

View synthesis process

Stereo

matching

SLIDE 14

View synthesis process

Stereo

matching

Disparity

computation

SLIDE 15

View synthesis process

Stereo

matching

Disparity

computation

Warping

SLIDE 16

View synthesis process

Stereo

matching

Disparity

computation

Warping

b: distance between the cameras D: disparity map

SLIDE 17

View synthesis process

Stereo

matching

Disparity

computation

Warping
Occlusion

handling

SLIDE 18

View synthesis with partial occlusion

SLIDE 19

Challenges: hole handling

Input images Hole handling Source: middlebury.edu

SLIDE 20

View synthesis with partial occlusion

SLIDE 21

View synthesis with partial occlusion

SLIDE 22

View synthesis with full occlusion

SLIDE 23

View synthesis with full occlusion

SLIDE 24

View synthesis with full occlusion

SLIDE 25

Anonymity

SLIDE 26

Interpolation

SLIDE 27

So, does it work?

nagaraja@iiitd.ac.in

SLIDE 28

Artifacts

SLIDE 29

Artifacts

SLIDE 30

Conclusions

We have posed a new privacy problem:

camera-location anonymity

We propose view-synthesis as a defense

technique to improve camera-location anonymity and photographer anonymity

Analytical results
Preliminary results show some promise

SLIDE 31

Questions?

nagaraja@iiitd.ac.in