Instructor: Peter Asaro asarop AT
newschool.edu
Teaching Assistant: Jill Shah shahj225 AT
newschool.edu
UAV FAA Consultant:
Course Numbers: NMDS 5283 CRN 7332
Time: Monday,
7:00 - 9:40 pm
Location: 6 East
16th St., Room 603
Course webpage is here: http://peterasaro.org/courses/2020Studio.html
Course blog is here: http://robotstudio2020.wordpress.com
This course explores the potential of robotic media platforms and computer vision for cinematic expression. As a Co-Lab, students will work in collaborative groups that will utilize the latest robotic and computer vision technologies to make short films. The first half of the semester will consist of an introduction to these technologies and in-class group exercises that will familiarize you with advanced digital camera techniques, and robotic camera control. These camera techniques and platforms will include advanced computer vision techniques such as Time-lapse, High Dynamic Range Imagery, Motion Magnification, Facial Recognition, Object Tracking, Optic Flow, and others, as well as 3D active-vision systems such as the Xbox Kinect. Robotic camera control will be explored through the use of remote-operated and computer-controlled servo-driven cameras, including RC vehicles, mobile robot dollys, robotic arms, and quadrotors (drones). We will explore a variety of control methods from remote control to pre-programmed and 3D model-driven control, as well as how these can be combined with vision techniques for the interactive control of cameras with gestures. We will also explore how these cinematographic techniques relate to visual storytelling and expression. In the second half of the course, students will pursue projects of their own design in groups, with the goal of producing a short experimental or narrative video utilizing these techniques. Previous programming experience is not required, but students will be expected to learn and apply basic programming skills in this course, and will be introduced to programming languages such as Processing, Python and Java, and programming platforms and libraries such as Arduino, ROS and OpenCV.
Please email me to setup an appointment.
You are expected to have thoroughly and thoughtfully read the assigned texts and to have prepared yourself to contribute meaningfully to the class discussions. Your participation will be evaluated in terms of both quantity and quality.
As this is a studio course, and a co-lab, regular attendance is essential. Because you will be doing group projects, if you do not come to class your fellow group members will be at a disadvantage. You will be permitted two excused absences (you must notify me and your group partners of your inability to attend before class, via email or phone). Any subsequent absences and any un-excused absences will adversely affect your grade.
For the first half of the class, studio time will be devoted to in-class group exercises that are designed to teach you fundamental concepts, techniques and how to apply these towards an aesthetic goal. Your results for each exercise should be posted to the course blog, along with a description. Generally, this will be a short video, program/software, or both. You should also note the names of the group members, and their roles, in your description. This should be done by the end of each class, but you may edit or update it later. You should also create or own an account on YouTube or Vimeo to uplaod any videos that you produce.
You will be required to create an account on WordPress, and send me an email with the EMAIL ADDRESS used to create the account, so that you can be added as authors for the collective course blog. Everyone will be posting to a common blog page, and this will be publicly accessible. When writing and making comments, you are expected to treat other students with the same respect and courtesy as you should in the classroom. You are also expected to respect rules of academic integrity, research ethics, and copyright when posting to the blog.
At the start of each class, we will review questions and concerns from the previous week, as well as review and critique the films that you have produced.
Proposals Due: April 17
Paper Due: May 15
There will be no final exam. Instead, your Final Project will be due, May 13th at 7:00PM. If that time will not work for you, you need to make other arrangements by May 4th at the latest.
The Final Project could take different forms, but should contain two key elements: robots and video/media. Final Projects will be evaluated in terms of technological and aesthetic innovation and quality. Ideally, aesthetics goals should drive the technology.
Some examples of good Final Projects:
Use of robotic platforms and/or computer vision for a short film
Use of robots in a short film
Development of advanced tools/technology for robotic/computational filmmaking
Development of an interactive media project that employs robotics and/or computer vision
A research paper that explores some advanced aspect of robotic media
Check out previous class projects here:
http://robotstudio2018.wordpress.com
http://robotstudio2017.wordpress.com
http://robotstudio2014.wordpress.com
In addition to submitting the final video clip, software and/or hardware that you created, you will also be asked to write-up an extended blog entry or webpage that describes how you created and used technologies to create your final project. These will be posted to the course blog. There may also be opportunities to submit films to festivals and events..
Because we have limited number of various robotics technologies and computers, and in an effort to build teamwork early, all of the studio projects will be done in groups of 3 or 4 people (depending on the number of students). On the first day of class, you will be asked to fill out a questionnaire to identify your skills and interests. You will then be put in a group which complements those skills. The goal is to have at least one person with film/video experience in each group, and at least one person with some programming experience in each group.
You will be in the same group from all of the in-class exerciese over the first 10 weeks. If one or more people miss a class, we may adjust the groups as needed.
You are free to form you own groups for the Final Project, or to continue with your exercise group.
We will need fast computers for the in-class exercises. You may use your own, or there will be Mac laptops available for checkout form the equipment center each week. Depending on the number of students, each group will get at least 1 laptop. Some exercises can only be done on a PC, and we will bring PCs for that, but most will work on the University Macs.
Equipment will be signed out at the beginning of class, and signed back in at the end, during the period of exercises. After that, or by special arrangement, equipment can be checkout out for longer periods to complete the Final Project outside of class.
We will be programming real robots and drones which move around in the world. They are mostly small and safe, but you should always use care and caution when working with them to protect yourself, classmates, and the public.
Don't run code if you are unsure of what it will do, or if you think it might be hazardous!
Anyone who behaves recklessly or endangers others will not be allowed to work with the robots any more.
We will review more detailed safety considerations when we start flying the quadrotors.
All readings will be available electronically, via the web, in PDF, MS Word, HTML, or similar format.
Most of the films and TV programs that will be assigned are available from a variety of sources. Many are available through the New School Library on DVD. In addition, they can be purchased from most book or video stores, rented from most video shop, or found through Netflix. For the videos which cannot be obtained easily in these ways, other means will be provided for you to view these films prior to class.
Grading StandardsCourse and Syllabus Overview
Questionnaire
Watch: Robot Media Videos:
https://www.youtube.com/watch?v=zINbyGTOFls
Behind the Scenes: Bot & Dolly "Box"
Sun Yuan & Peng Yu's "Can't Help Myself"
Behind the Scenes: RC Car chase
OK Go "I Won't Let You Down" music video
Recommended Additional Readings:
Bill
Gates (2007) "A Robot in Every Home: The leader of the PC revolution predicts
that the next hot field will be robotics," Scientific American, January
2007.
Hans
Moravec (2009) "Rise of the Robots--The Future of Artificial Intelligence,"
Scientific American, March 23, 2009
How Digital Cameras Work How Digital Cameras Work on Astropix
The History and Science of Lenses
More about Camera Hacking
OpenCV
Processing
Canny
Edge Dector in OpenCV
https://github.com/atduskgreg/opencv-processing
(should be installed)
http://urbanhonking.com/ideasfordozens/2013/07/10/announcing-opencv-for-processing/
Recommended Additional Readings:
Watch: NVIDIA Drive Net Demo
Watch: NVIDIA Drive
PX2 self-driving car
Listen: http://www.radiolab.org/story/211119-colors/
OpenCV http://opencv.org/
http://en.wikipedia.org/wiki/High_dynamic_range_imaging
http://en.wikipedia.org/wiki/Time-lapse_photography
Intro to HDR (1 hour 30 min) http://www.youtube.com/watch?v=v3CPavb2NWs
HDR with RAW file in Photoshop http://www.youtube.com/watch?v=YY7cYyXjQ-I
http://en.wikipedia.org/wiki/CCD_camera
http://en.wikipedia.org/wiki/Active_pixel_sensor
http://en.wikipedia.org/wiki/Digital_photography
Watch: https://vimeo.com/27582408
http://en.wikipedia.org/wiki/Camera_lens
http://photo.net/learn/optics/lensTutorial
Kinect and Processing Tutorials
The Depth Image in Kinect and Processing
Recommended Further Readings:
Open NI https://en.wikipedia.org/wiki/OpenNI Open NI on Githubhttps://github.com/OpenNI/OpenNI
J. J. Gibson https://en.wikipedia.org/wiki/James_J._Gibson
Time-of-Flight Camera http://en.wikipedia.org/wiki/Time-of-flight_camera
Bullet Time http://en.wikipedia.org/wiki/Bullet_time
Free Viewpoint Television http://en.wikipedia.org/wiki/Free_viewpoint_television
Watch (in class):
The Science of Rendering Photorealistic CGI
Point Clouds in Kinect and Processing
Hand tracking in Kinect and Processing
Recommended:
http://www.pointclouds.org/documentation/tutorials/using_kinfu_large_scale.php Prior to class:
If you have an iPhone, iPad or Android device,
please download and install this app from iTunes, GooglePlay or the Android
Market: AR.FreeFlight https://itunes.apple.com/us/app/freeflight/id373065271?mt=8 It is also strongly recommended that you download
this simulator, and practice with it before class: ARDrone Sim https://play.google.com/store/apps/details?id=com.DDI.ARDroneSim&hl=en For the Simulator options choose: AR.Drone 2.0, Indoor Hull, Standard Battery Press the Green Takeoff button to launch and
hover, the Green Landing button for an automatic controlled landing. The Left joystick tilts the drone, and causes it
to move laterally. The Right joystick causes the drone to go up and
down, and rotate clockwise or counter clockwise. The Camera icon will shift your viewing
perspective. If you crash, it will simulate damage, so you may
need to go to Settings on the main screen and Restore Defaults to fly
straight again. Studio: FAA Drone Regulations Where to Fly Drones in NYC Watch: Inside Google's
"Daydreaming" Computer Neural Network that Changes
Everything Maroon 5, "Cold ft. Future"
music video Make your own DeepDream
Image (upload) Other
Resources: Watch: How to Make a Simple Tensorflow Speech Recognizer Links if you want to try TensorFlow: Installing TFLearn, TensorFlow, Python Links from the lecture: Using Wolfram|Alpha for 3D plots RoboScience Robot Simulator "Veranda" Learning the LOGO Programming Language ACLU trains Amazon Rekognition to recognize criminals, and uses it on Congress Watch before class: Intro to Computer Vision in Processing (23 minutes) Computer Vision: Motion Detection (17 minutes) Required: Oscar's
Slide Presentation 2018 Oscar's
Slide Presentation 2017 15 Tips for Drone
Cinematography Recommended Further
Readings: Crane & Aerial shots Chayka,
Kyle "The Troubling Contradictions of Dronestragrams," The New
Republic, April 19. 2017. Belton, John. "The Bionic Eye: Zoom
Esthetics." Cineaste (1980): 20-27. Recommended Further
Readings: Required:
Microsoft Robotics Studio
Willow
Garage ROS (Robot Operating System)
ROS & Blender
http://www.openrobots.org/morse/doc/stable/morse.html
http://opengrasp.sourceforge.net
ROS http://www.ros.org/wiki/
ROS & Arduino http://wiki.ros.org/rosserial
http://www.ros.org/news/2013/03/new-package-ros-arduino-bridge.html
http://wiki.ros.org/arbotix
Watch:
Craig Gillespie, Lars and the
Real Girl, 2007, 106 min
Watch: David
Hanson: Robots that "show emotion", 2010 TED talk, 5 min.
Watch: Caleb Chung plays with Pleo, 2010 TED talk, 17 min.
Watch: Leila Takayama, "What's it like to be a robot?", 2017 TED talk, 13 min.
Watch: Kate Darling, "Why we have an emotional connection to robots,"
2018 TED talk, 12 min.
Part II: Actuation
Week 5: March 9
Drones I: Ground School
http://ardrone2.parrot.com/apps/
Note: This app costs $1.99 (iOS) or $2.39
(Android)
Week of March 16: NO CLASS! SPRING BREAK
Week 6: March 23
Dreaming Like a RobotStudio Assignment:
Studio 5:
Using Google DeepDream
http://www.ros.org/news/2013/03/new-package-ros-arduino-bridge.html
http://wiki.ros.org/arbotix
Week 7: March 30
More Neural NetworksStudio Assignment (Revised):
Studio 5:
Using Google DeepDream
Week 8: April 6
More Computer Vision in Processing and OpenCV
Final Project Proposals Due!
Studio Assignment:
Studio 6:
OpenCV (in Processing) for Face Filters & Motion Detection
Week 9: April 13
Drones Cinematography Techhnues with DJI
Sparks
Zoom Guest Lecture: Oscar Frasser oscarfrasser@newschool.edu
Paul
Virilio, "Cinema isn't I See, it's I Fly," in War and Cinema: The
Logistics of Perception, London: Verso, 1989: 11-30.
Beth Herst, "Review:
The Disembodied Eye," PAJ: A Journal of Performance and Art, Vol. 24,
No. 1, Intelligent Stages: Digital Art
and Performance (Jan., 2002), pp. 122-126.Week 10: April 20
Virtual & Augmented Reality
Virtual Field Trip: XReality Center with Director Maya Georgieva
georm664@newschool.eduWeek 11: April 27
Guest Lecture: Nina & Georgi Tushev, Tushev's Aerials
Week 12: May 4
Drones & Blimp Drones (meet in regular
classroom)
Guest Lecture: Will Chatham
Week 13: May 11
Final Project PresentationsWeek 14: May 13
NO CLASS MEETING
Final Projects Due