Who’s there?

More than a spot for the Alia2 Foundation

How do you visualize the lack of protection from criminals who find disguise behind fake web identities?

By allowing people to associate with a fragile, vulnerable child and then uncovering the truth that could lie behind a person’s internet identity. This surprises the viewer and makes them feel shocked and disgusted by the criminal intentions.

Alia2 - Spot

Client

Alia2 is a Spanish non-profit organization that works to protect the rights of children on the internet by promoting its safe and responsible use among the public sector, the private sector and society. We implemented this project in collaboration with BBDO Spain.

Many people don’t know the meaning of the word child grooming, and even less, that it’s the third most lucrative felony in the world. Miguel Comín, cofounder of Alia2

Challenge

On an unregulated internet, pedophiles can easily take the identity of whoever they choose, including a child. Alia2 was looking for an eye-opening spot to stimulate viewers’ empathy for children who are vulnerable to the internet’s lack of security.

Idea

The viewer follows a chat conversation between two young girls. He can see one of them, but only as an avatar. Gradually, it becomes clear that there is something wrong with her identity and finally, a modification of her voice and a change in the camera’s perspective reveal that she in fact not a girl but a grown man.

We saw Kinect was an ideal solution to achieve our goal. Its data is a single side pointcloud, just like a mask onto which you can project images. What may seem as a limitation of this technology ended up being a perfect fit for this project.

  • Idea

Aesthetics

Designing the subject with a grid of lines introduces the idea of a questionable identity.

The depth camera used our base data to provide a high number pointcloud. Different visual solutions were considered in order to obtain the desired look. In order to show a fragile, vulnerable subject, a young girl with a questionable identity, we combined three layers: pointcloud, wireframe and mesh. Each layer had different light sources, materials and shades. We also applied cinematographic effects like depth of field and bloom in order to achieve a render style look.

  • Pointcloud: Kinect Point cloud
    Kinect Point cloud
  • Pointcloud: Kinect Point cloud
    Mesh wireframe
  • Pointcloud: Kinect Point cloud
    Mesh
  • Pointcloud: Kinect Point cloud
    Merged layers

Technology

Kinect - Depth sensing

The depth sensor consisted of an infrared laser projector combined with a monochrome sensor, which can capture video data in 3D under any ambient light condition. Each frame of the depth data stream was made up of pixels that contained the distance from the camera plane to the nearest object.

Kinect

Pointcloud to mesh converter

  • Point Cloud
    Point Cloud
  • Point Mesh
    Point Mesh

Custom OpenGL
+ Openframeworks application

The application has three modes: recording, editing and rendering. In recording mode we stored all the information provided by the Kinect in a raw text file on a separated thread.

Custom OpenGL

Timeline animation system

Our team developed an interface based on ofxTimeline where we could edit the recorded content and the environment: lighting, shadows, positions and sizes separately for the different visualization layers (pointcloud, wireframe and mesh).

Timeline animation system