1. DRAWING APPRENTICE
BY ANAMIKA SHARAF

2. What am I?

3. Continue.. Enactive AI Drawing partner
Reciprocal feedback loop between user and the system.
Human-AI pair as a creative system
Web based Drawing application
Created by Adaptive Digital Media Lab [ ADAM Lab]

4. Let’s Watch https://www.youtube.com/watch?v=D72y-9wNzxo

5. Few interesting art created

6. Continue..
User’s lines black and AI agent’s lines are blue.

7. Continue..

8. Bridging the gap between CST and Computational Creativity

9. Time lapsed representation of Picasso’s abstract art in Picasso painting in Clouzot(1956)

10. Approaches Some approaches that has yielded interesting examples are:
Mimicry
Structured improvisation
Using contextual clues to negotiate shared mental models.
For example:
The improvisational percussion robot Shimon mimics human musicians by analyzing the rhythm and pitch of musical performances and generating synchronized melodic improvisations (Hoffman & Wein- berg 2010).
Call and response interaction where each party modifies and build on each other’s contribution.

11. Enactive Cognition
It is an interaction between an acting organism and it’s environment.
Our environment is one which we selectively create through our capacities to interact with the world.

12. Enactive Model

13. Continue..
Routine actions only require minimal thought and a limited amount of highly relevant sensory data.
small deviations to the left to update and revise strategy, and deviations to the right to interactively evaluate those ideas in a perceive-act cycle
if the agent is performing an unfamiliar task, however, cognitive resources are recruited to actively build a mental model of the situation, which requires performing experimental interactions, closely examining the results in the environment, and then updating the mental model in a slower perceive-think-act cycle.

14. Perceptual Logic
Perceptual logic is their proposed method for developing ‘intelligent’ perception in an agent.
Perceptual logic is a proposed cognitive mechanism that filters sensory data, identifies relevant percept-action pairings, and presents these percept-action pairings as affordances to perception. Perceptual logic performs a similar role as the ‘simulator’ in Perceptual Symbol Systems (Barsalou 1999).
The simulator activates all the associated information related to a percept, including the various ways it can be interacted with based on experiential knowledge and physical characteristics.

15. Drawing Apprentice Software Architecture

16. Continue.. The creative dialogue begins as the human inputs a line
All current lines from the canvas are sent to the perceptual logic module.
The perceptual logic module consults the creative trajectory monitor to determine what perceptual logic to apply to its current data set.
The planned creative trajectory monitor has a coarse grained record of the previous drawing behavior based on the time between the user’s lines (i.e. longer periods of rest represent reflection, which is categorized as global perceptual logic, and short and rapid detail strokes are categorized as local perceptual logic).

17. Layers of Perceptual Logic

18. EMC suggests that each layer of perceptual logic should generate unique artistic affordances from the same input, such as shading a circle, intersecting it, and replicating it.
Each logic layer sends its algorithms different amount of lines and different features for discriminating lines.
There are several critical points that each perceptual logic filter can use in different ways, such as inflection points, start point, end point, segments between inflections, and corners. Moreover, gestalt groupings (e.g. proximity, similarity, closure, etc.) provide additional features to generate unique affordances building relationships between lines, groups of lines, regions, and patterns

20. Features Draws with user in real time
Analyze user’s input lines and responds with lines of it’s own.
An investigation in this new domain of human-computer collaboration such as methods of feedback to facilitate learning, coordination( for both), the role control and ambiguity plays in effective collaborations.

21. Recognition
ACM Digital Library has awarded Drawing Apprentice in code based art category.
Evaluation was based on formative user studies and expert evaluation.

22. References [1] Adam Lab: http://adamlab.gatech.edu/?page_id=711
[2] ACM Digital Library:
[3] Enactivism:
[4]Building artistic Compter Colleagues with an Enactive Model of creativity: