changeset 521:2529c34caa1a

changes from mom.
author rlm
date Mon, 31 Mar 2014 08:29:50 -0400
parents 1803144ec9ae
children 1e51263afdc0
files thesis/cortex.bib thesis/cortex.org
diffstat 2 files changed, 13 insertions(+), 14 deletions(-) [+]
line wrap: on
line diff
     1.1 --- a/thesis/cortex.bib	Mon Mar 31 08:21:39 2014 -0400
     1.2 +++ b/thesis/cortex.bib	Mon Mar 31 08:29:50 2014 -0400
     1.3 @@ -114,7 +114,7 @@
     1.4                    embodiment is critical to intelligence.}}
     1.5  } 
     1.6  
     1.7 -@book{9.01-textbook,
     1.8 +@book{textbook901,
     1.9    author="Bear and Mark F. and Barry W. Connors and Michael A.",
    1.10    title="Neuroscience: Exploring the Brain.",
    1.11    publisher="Lippincott Williams \& Wilkins",
     2.1 --- a/thesis/cortex.org	Mon Mar 31 08:21:39 2014 -0400
     2.2 +++ b/thesis/cortex.org	Mon Mar 31 08:29:50 2014 -0400
     2.3 @@ -186,7 +186,7 @@
     2.4     model of your body, and aligns the model with the video. Then, you
     2.5     need a /recognizer/, which uses the aligned model to interpret the
     2.6     action. The power in this method lies in the fact that you describe
     2.7 -   all actions form a body-centered viewpoint. You are less tied to
     2.8 +   all actions from a body-centered viewpoint. You are less tied to
     2.9     the particulars of any visual representation of the actions. If you
    2.10     teach the system what ``running'' is, and you have a good enough
    2.11     aligner, the system will from then on be able to recognize running
    2.12 @@ -296,10 +296,10 @@
    2.13     
    2.14  *** Main Results 
    2.15  
    2.16 -   - After one-shot supervised training, =EMPATH= was able recognize a
    2.17 -     wide variety of static poses and dynamic actions---ranging from
    2.18 -     curling in a circle to wiggling with a particular frequency ---
    2.19 -     with 95\% accuracy.
    2.20 +   - After one-shot supervised training, =EMPATH= was able to
    2.21 +     recognize a wide variety of static poses and dynamic
    2.22 +     actions---ranging from curling in a circle to wiggling with a
    2.23 +     particular frequency --- with 95\% accuracy.
    2.24  
    2.25     - These results were completely independent of viewing angle
    2.26       because the underlying body-centered language fundamentally is
    2.27 @@ -381,7 +381,7 @@
    2.28     #+ATTR_LaTeX: :width 12cm
    2.29     [[./images/blender-worm.png]]
    2.30  
    2.31 -   Here are some thing I anticipate that =CORTEX= might be used for:
    2.32 +   Here are some things I anticipate that =CORTEX= might be used for:
    2.33  
    2.34     - exploring new ideas about sensory integration
    2.35     - distributed communication among swarm creatures
    2.36 @@ -542,7 +542,7 @@
    2.37     Most human senses consist of many discrete sensors of various
    2.38     properties distributed along a surface at various densities. For
    2.39     skin, it is Pacinian corpuscles, Meissner's corpuscles, Merkel's
    2.40 -   disks, and Ruffini's endings (\cite{9.01-textbook), which detect
    2.41 +   disks, and Ruffini's endings \cite{textbook901}, which detect
    2.42     pressure and vibration of various intensities. For ears, it is the
    2.43     stereocilia distributed along the basilar membrane inside the
    2.44     cochlea; each one is sensitive to a slightly different frequency of
    2.45 @@ -560,8 +560,8 @@
    2.46     each sense. 
    2.47  
    2.48     Fortunately this idea is already a well known computer graphics
    2.49 -   technique called called /UV-mapping/. The three-dimensional surface
    2.50 -   of a model is cut and smooshed until it fits on a two-dimensional
    2.51 +   technique called /UV-mapping/. The three-dimensional surface of a
    2.52 +   model is cut and smooshed until it fits on a two-dimensional
    2.53     image. You paint whatever you want on that image, and when the
    2.54     three-dimensional shape is rendered in a game the smooshing and
    2.55     cutting is reversed and the image appears on the three-dimensional
    2.56 @@ -651,10 +651,9 @@
    2.57          pipeline. The engine was not built to serve any particular
    2.58          game but is instead meant to be used for any 3D game. 
    2.59  
    2.60 -   I chose jMonkeyEngine3 because it because it had the most features
    2.61 -   out of all the free projects I looked at, and because I could then
    2.62 -   write my code in clojure, an implementation of =LISP= that runs on
    2.63 -   the JVM.
    2.64 +   I chose jMonkeyEngine3 because it had the most features out of all
    2.65 +   the free projects I looked at, and because I could then write my
    2.66 +   code in clojure, an implementation of =LISP= that runs on the JVM.
    2.67  
    2.68  ** =CORTEX= uses Blender to create creature models
    2.69