Mercurial > cortex
view thesis/rlm-cortex-meng.tex @ 427:c670d23003de
rename.
author | Robert McIntyre <rlm@mit.edu> |
---|---|
date | Fri, 21 Mar 2014 15:36:37 -0400 |
parents | |
children | d53a31969a51 |
line wrap: on
line source
2 \section{Artificial Imagination}3 \label{sec-1}5 Imagine watching a video of someone skateboarding. When you watch6 the video, you can imagine yourself skateboarding, and your7 knowledge of the human body and its dynamics guides your8 interpretation of the scene. For example, even if the skateboarder9 is partially occluded, you can infer the positions of his arms and10 body from your own knowledge of how your body would be positioned if11 you were skateboarding. If the skateboarder suffers an accident, you12 wince in sympathy, imagining the pain your own body would experience13 if it were in the same situation. This empathy with other people14 guides our understanding of whatever they are doing because it is a15 powerful constraint on what is probable and possible. In order to16 make use of this powerful empathy constraint, I need a system that17 can generate and make sense of sensory data from the many different18 senses that humans possess. The two key proprieties of such a system19 are \emph{embodiment} and \emph{imagination}.21 \subsection{What is imagination?}22 \label{sec-1-1}24 One kind of imagination is \emph{sympathetic} imagination: you imagine25 yourself in the position of something/someone you are26 observing. This type of imagination comes into play when you follow27 along visually when watching someone perform actions, or when you28 sympathetically grimace when someone hurts themselves. This type of29 imagination uses the constraints you have learned about your own30 body to highly constrain the possibilities in whatever you are31 seeing. It uses all your senses to including your senses of touch,32 proprioception, etc. Humans are flexible when it comes to "putting33 themselves in another's shoes," and can sympathetically understand34 not only other humans, but entities ranging from animals to cartoon35 characters to \href{http://www.youtube.com/watch?v=0jz4HcwTQmU}{single dots} on a screen!38 \begin{figure}[htb]39 \centering40 \includegraphics[width=5cm]{./images/cat-drinking.jpg}41 \caption{A cat drinking some water. Identifying this action is beyond the state of the art for computers.}42 \end{figure}45 This is a basic test for the vision system. It only tests the46 vision-pipeline and does not deal with loading eyes from a blender47 file. The code creates two videos of the same rotating cube from48 different angles.51 \begin{clojurecode}52 (in-ns 'cortex.test.vision)54 (defn test-pipeline55 "Testing vision:56 Tests the vision system by creating two views of the same rotating57 object from different angles and displaying both of those views in58 JFrames.60 You should see a rotating cube, and two windows,61 each displaying a different view of the cube."62 ([] (test-pipeline false))63 ([record?]64 (let [candy65 (box 1 1 1 :physical? false :color ColorRGBA/Blue)]66 (world67 (doto (Node.)68 (.attachChild candy))69 {}70 (fn [world]71 (let [cam (.clone (.getCamera world))72 width (.getWidth cam)73 height (.getHeight cam)]74 (add-camera! world cam75 (comp76 (view-image77 (if record?78 (File. "/home/r/proj/cortex/render/vision/1")))79 BufferedImage!))80 (add-camera! world81 (doto (.clone cam)82 (.setLocation (Vector3f. -10 0 0))83 (.lookAt Vector3f/ZERO Vector3f/UNIT_Y))84 (comp85 (view-image86 (if record?87 (File. "/home/r/proj/cortex/render/vision/2")))88 BufferedImage!))89 (let [timer (IsoTimer. 60)]90 (.setTimer world timer)91 (display-dilated-time world timer))92 ;; This is here to restore the main view93 ;; after the other views have completed processing94 (add-camera! world (.getCamera world) no-op)))95 (fn [world tpf]96 (.rotate candy (* tpf 0.2) 0 0))))))97 \end{clojurecode}100 \begin{itemize}101 \item This is test1 \cite{Tappert77}.102 \end{itemize}