FMP 13. FMP Project 03

Afterwards, I felt this is not enough when it comes to the method of showing saved images because people will not know the order of images, namely the time people or me are captured. It is highly possible to just look at looping images blankly without any functions. In the sense, I decided to show it with a ‘MATRIX’ (x, y coordinate) of images ‘CHRONOLOGICALLY’ with the most recent images approximately 30-50. It might help people see the previous events and people as well as the captured oneself efficiently. This is a matrix of images deployed 6 by 6 chronologically.

screen-shot-2016-11-25-at-8-43-51-pm screen-shot-2016-11-25-at-8-44-03-pm

However, it still have a lot of time to look at my turn. People have to wait maximum 36 second to check their picture, and they might be just passing through without any idea of this. Therefore, I have to reduce the time appearing pictures. As the compromise of the way showing images, images deployed in sequence just pop up at the same time when people are captured. It can be a good way to help people to check their pictures effectively without just passing though the screen. And this is the final outcome of my project process, and using the programme. I am going to show my project based on my theory to people effectively.


And, in order to organise and show the process of the development of my working process, I made a video.


FMP 12. FMP Project 02

However, in terms of using video, if there are no people in front of webcam, people cannot experience the blurred faces. There is nothing when people look at the screen. In order to overcome the downside regarding the feature of video, I decided to make image-based surveillance, not video-based. It means that when people pass in front of a webcam, the cam recognises and captures faces. After capturing, the captured images are automatically saved, and the stored pictures are loaded on the screen afterwards. It can make people to experience both of aspects in the space. To sum it up, before coming to the screen, they are captured with blurred faces, and then when they get in front of the screen, they realise that they are confidentially captured. However, due to the face protection system for visual anonymity, their identity is not exposed as a point of compromise. And, it is the most effective way to show my theory. This is the first step regarding capturing and saving pictures.


And then, recalling the saved images on the screen in real time is the second step.

screen-shot-2016-11-27-at-8-45-21-am screen-shot-2016-11-27-at-8-47-57-am

FMP 11. FMP Project 01

I want visitors to experience ‘camera surveillance’ and ‘face protection for the camera surveillance for visual anonymity’ which are the core value of my project. I sketched a simple expectation of the outcome of my work in the postgraduate show. Through a ‘webcam representing a drone camera’ and ‘blurring faces’ with real-time face protection programme’, people can experience a surveillance camera and face protection in a way.

Idea Sketch

For this planning, by using OpenCV and Processing, I can make a real-time face projection system by detecting and blurring faces as well. I Found out several materials to hide faces in real time.

Here is my test video using OpenCV and Processing with codes.

screen-shot-2016-11-29-at-2-34-58-am screen-shot-2016-11-29-at-2-35-20-am

It is successful to conduct ‘camera surveillance’ and ‘face protection’ at the same time. However, the problem is that in front of the screen and webcam, it is ‘NOT’ surveillance like CCTVs. It is more like just a experience blurred faces. For experiencing camera surveillance system, the key word are ‘confidential’ and ‘indiscriminate’ which mean that without permission, a cam is shooting people, and then when people come to the screen, they show their blurred pictures accidentally.

FMP 10. Documentary Film 02

After shooting sources, I designed a basic interface design of a drone’s eyes based on the essential elements for surveillance camera such as the machine’s name, time code, date, location where the drone is and etc. The main issue of this interface is that several information such as time code, date and live face protection should be working in real time.

Basic interface of a drone’s eyes

Afterwards, I made an image-based scenario of the documentary film in sequence before starting making a video in advance.

screen-shot-2016-11-28-at-3-02-48-pm screen-shot-2016-11-28-at-3-03-25-pm screen-shot-2016-11-28-at-3-03-47-pm screen-shot-2016-11-28-at-3-04-01-pm screen-shot-2016-11-28-at-3-04-21-pm screen-shot-2016-11-28-at-3-04-39-pm screen-shot-2016-11-28-at-3-08-33-pm screen-shot-2016-11-28-at-3-09-27-pm screen-shot-2016-11-28-at-3-09-48-pm screen-shot-2016-11-28-at-3-10-05-pm screen-shot-2016-11-28-at-3-10-17-pm screen-shot-2016-11-28-at-3-10-32-pm screen-shot-2016-11-28-at-3-10-44-pm screen-shot-2016-11-28-at-3-11-04-pm screen-shot-2016-11-28-at-3-11-19-pm screen-shot-2016-11-28-at-3-12-08-pm screen-shot-2016-11-28-at-3-12-25-pm

The below is the final documentary film I made.

By making a documentary film, I have the opportunity to focus on the right direction of my theory to design something, and I got a proper research material to enable people to understand my thoughts and project effectively.

FMP 09. Documentary Film 01

I was using a drone to made a documentary film to help people to effectively understand my theory in terms of ‘the era of surveillance drone policing’ and ‘face protection for privacy’ by adding ‘the scenes of drone patrols’, ‘the effect of blurring faces’. Besides, by putting some text that can support my theory regarding research and quotes into the film, I am looking forward to the film being made as an efficient communication tool.

Parrot Bebop 2

A scene during shooting

I made testing films in order to test the way of showing situations briefly The first is pixelating faces during Surveillance Drone Policing to protect ‘visual anonymity’, and the second is shooting a wide range of blind spots with many cuts to make the situation of drone patrols more dramatic.

FMP 08. What if Question

Based on the previous researches, I have had a question about a possibility like below.

With real-time facial recognition system on drones’s eyes, what if surveillance drone policing system itself may ‘protect’ privacy for people’s visual anonymity by blurring faces? is it possible to become points of compromise of tensions between privacy and surveillance?

Based on the question, I made a project title and description of it.

Title – Point of Compromises (Between ‘Privacy’ for Visual Anonymity and ‘Surveillance’ for Security)

Description – Points of Compromise proposes a rational system design for surveillance drone policing. It explores how camera monitoring systems may ‘protect’ the public. The project explores the tensions between privacy and surveillance.

FMP 07. Protection system as a Form of Compromise

As a form of ‘Compromise’, not like a form of ‘Resistance’ like ‘CV Dazzle’, two typical corporations: YouTube and Google have had endeavours to protect faces. They launched a function to blur faces to keep ‘Visual Anonymity’.

In 2012, YouTube uploaded a post with a title ‘Face blurring: when footage requires anonymity’. Some sentences inform us of their key aim of the service.

“Whether you want to share sensitive protest footage without exposing the faces of the activists involved, or share the winning point in your 8-year-old’s basketball game without broadcasting the children’s faces to the world, our face blurring technology is a first step towards providing visual anonymity for video on YouTube.”

“Visual anonymity in video allows people to share personal footage more widely and to speak out when they otherwise may not.”

“Because human rights footage, in particular, opens up new risks to the people posting videos and to those filmed, it’s important to keep in mind other ways to protect yourself and the people in your videos.”

These systems show us how privacy and personal identity are significant.

Google Street View