In this project we wanted to recreate an artificial long sequence, as happened in Birdman, trying to use both physical transition and post production VFX.
The main limit was to use only few instruments. In fact we had:
- Canon 600d
- Lens 50 mm fixed focal length optic
- Shoulder support
We started writing the script of the sequence. We take inspiration by Birdman story, but we have to condense it in few minutes. The main character is an actress that have just finished her performance and moves from the stage to the back of the theatre meeting people and crossing doors.
We wanted to use an ambient that was similar to the theatre that appears in Birdman, so we succeed in filming at CAP10100 theatre in Turin, using all the room that we can access to.
Then we divided the sequence in 7 parts. Each part is a single take and the transition between those parts had to be smooth to mask the fact that were different takes.
There are lot of possible camera transitions that can be used for this goal. We selected some of them that can be useful for our story.
The 7 break are so divided:
1. The camera is in the audience and look at the actress (A1) in the center of the stage. The camera moves (red line) to the actor and stop when it reach the beginning of the stage (here there’s a gap about 1,5 m tall).
2. The camera is on the stage, moves around the actress (A1), on the foreground appear a second actor (A2) that cross the audience (blue line), the camera start following him till he reach the end of the stage were the curtains mask the camera.
3. The camera is at the beginning of a corridor, it’s hidden by a black door that opens when the camera moves. The camera follows the second actor (A2) till he enter a door closing it before the camera can pass. At this point the camera makes a whip movement to the wall.
4. The camera is at the beginning of another corridor and it’s fixed. Then appear the first actor (A1) from the back of the camera and walks through the corridor. The camera follow him till he enter another room. The camera does not orientate itself according to actor movement and enter the room with an horizontal movement.
5. The camera enter the room horizontally and then rotate to reach the firs actor (A1). The the second actor (A2) enter the room and look at his telephone with the first actor (A1). On the telephone there’s a video of the performance of the first actor (A1). The camera moves closely to the telephone.
6. The camera moves back from the telephone and then look at the person standing in front of the bar. Then appear the first actor (A1) that walks near the bar. The camera moves horizontally filming the actor that walks and stop when he reach the end of the bar. There is a person standing at the corner that mask the camera.
7. The camera is outside the bar and begin from the person standing a the corner and then follow the first actor (A1) that stop at the bar and start drinking.
Then we passed at the post production process, in which we worked on the clips to make them match perfectly.
1. On the first take the transition was made by darkness so there wasn’t so much work to do. To match better the clip 1 and 2 we use some audio effects to simulate a crowd in a theatre.
2. Also the second take use a transition to black, but in a different way.
3. On the third take we use a whip movement, so in post production we work on match the first whip with the second. We worked on opacity and masks.
4. On the fourth take we used the jamb as a mask and we worked also with color correction effect and on the camera speed to matte the two clips.
5. On the fifth take it was very difficult to track the camera movement at the end of the clip because the focus length of the lens was not enough to make clear near object, so we used Mocha to match the video of the performance on the display of the smartphone. We changed also opacity, focus length, exposure, edge blur of the attached video to match with the changing focus of the camera. We focused some pixels of the screen and we use them to attached the next clip with the same process.
6. On the sixth take we use a mask on the person at the corner of the bar.
7. On the seventh take we use the same mask to connect the previous clip.
Then we made some color correction adjustments to uniform the clips and audio edition.