（7）face and 2-d mesh animation in mpeg-4.htm

关于MPRG4的一些基本的指南

with the expression prior to superimposing FAPs 3-68. This deformation is 
applied with the neutral face constraints of mouth closure, eye opening, gaze 
direction and head orientation. Since the encoder does not know how FAP 1 and 2 
are implemented, we recommend using only those low-level FAPs that will not 
interfere with FAP 1 and 2.</P>
<P align=justify>&nbsp;</P>
<P align=center><B><A name=_Ref416295458>Table 2</A>: FAP groups</B></P><B>
<DIV align=center>
<CENTER>
<TABLE cellSpacing=1 cellPadding=7 width=474 border=1>
  <TBODY>
  <TR>
    <TD vAlign=top width="62%"><FONT size=1>
      <P align=justify>Group</FONT></P></TD>
    <TD vAlign=top width="38%"><FONT size=1>
      <P align=center>Number of FAPs</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="62%"><FONT size=1>
      <P align=justify>1: visemes and expressions </FONT></P></TD>
    <TD vAlign=top width="38%"><FONT size=1>
      <P align=center>2</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="62%"><FONT size=1>
      <P align=justify>2: jaw, chin, inner lowerlip, cornerlips, midlip 
      </FONT></P></TD>
    <TD vAlign=top width="38%"><FONT size=1>
      <P align=center>16</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="62%"><FONT size=1>
      <P align=justify>3: eyeballs, pupils, eyelids </FONT></P></TD>
    <TD vAlign=top width="38%"><FONT size=1>
      <P align=center>12</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="62%"><FONT size=1>
      <P align=justify>4: eyebrow </FONT></P></TD>
    <TD vAlign=top width="38%"><FONT size=1>
      <P align=center>8</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="62%"><FONT size=1>
      <P align=justify>5: cheeks </FONT></P></TD>
    <TD vAlign=top width="38%"><FONT size=1>
      <P align=center>4</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="62%"><FONT size=1>
      <P align=justify>6: tongue</FONT></P></TD>
    <TD vAlign=top width="38%"><FONT size=1>
      <P align=center>5</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="62%"><FONT size=1>
      <P align=justify>7: head rotation </FONT></P></TD>
    <TD vAlign=top width="38%"><FONT size=1>
      <P align=center>3</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="62%"><FONT size=1>
      <P align=justify>8: outer lip positions</FONT></P></TD>
    <TD vAlign=top width="38%"><FONT size=1>
      <P align=center>10</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="62%"><FONT size=1>
      <P align=justify>9: nose</FONT></P></TD>
    <TD vAlign=top width="38%"><FONT size=1>
      <P align=center>4</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="62%"><FONT size=1>
      <P align=justify>10: ears </FONT></P></TD>
    <TD vAlign=top width="38%"><FONT size=1>
      <P align=center>4</FONT></P></TD></TR></TBODY></TABLE></CENTER></DIV>
<P align=justify><A name=_Ref416252498></A><A name=_Ref416785652></A><A 
name=_Ref416323649></A></P>
<P align=center><A name=_Ref435856395></A>Table 3: Visemes and related 
phonemes.</P>
<DIV align=center>
<CENTER>
<TABLE cellSpacing=1 cellPadding=7 width=342 border=1>
  <TBODY>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>Viseme #</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>phonemes</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify>example</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>0</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>none</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify>na</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>1</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>p, b, m</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify><U>p</U>ut, <U>b</U>ed, <U>m</U>ill</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>2</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>f, v</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify><U>f</U>ar, <U>v</U>oice</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>3</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>T,D</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify><U>th</U>ink, <U>th</U>at</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>4</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>t, d</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify><U>t</U>ip, <U>d</U>oll</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>5</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>k, g</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify><U>c</U>all, <U>g</U>as</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>6</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>tS, dZ, S</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify><U>ch</U>air, <U>j</U>oin, <U>sh</U>e</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>7</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>s, z</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify><U>s</U>ir, <U>z</U>eal</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>8</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>n, l</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify><U>l</U>ot, <U>n</U>ot</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>9</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>r</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify><U>r</U>ed</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>10</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>A:</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify>c<U>a</U>r</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>11</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>e</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify>b<U>e</U>d</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>12</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>I</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify>t<U>i</U>p</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>13</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>Q</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify>top</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="29%"><FONT size=1>
      <P align=justify>14</FONT></P></TD>
    <TD vAlign=top width="31%"><FONT size=1>
      <P align=justify>U</FONT></P></TD>
    <TD vAlign=top width="40%"><FONT size=1>
      <P align=justify>b<U>oo</U>k</FONT></P></TD></TR></TBODY></TABLE></CENTER></DIV>
<P align=center><A name=_Ref416295897>Table 4</A>: Primary facial expressions as 
defined for FAP 2.</P>
<TABLE cellSpacing=1 cellPadding=7 width=576 border=1>
  <TBODY>
  <TR>
    <TD vAlign=top width="7%"><FONT size=1>
      <P align=justify>#</FONT></P></TD>
    <TD vAlign=top width="32%"><FONT size=1>
      <P align=justify>Expression name</FONT></P></TD>
    <TD vAlign=top width="60%"><FONT size=1>
      <P align=justify>textual description</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="7%"><FONT size=1>
      <P align=justify>1</FONT></P></TD>
    <TD vAlign=top width="32%"><FONT size=1>
      <P align=justify>Joy</FONT></P></TD>
    <TD vAlign=top width="60%"><FONT size=1>
      <P align=justify>The eyebrows are relaxed. The mouth is open and the mouth 
      corners pulled back toward the ears.</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="7%"><FONT size=1>
      <P align=justify>2</FONT></P></TD>
    <TD vAlign=top width="32%"><FONT size=1>
      <P align=justify>Sadness</FONT></P></TD>
    <TD vAlign=top width="60%"><FONT size=1>
      <P align=justify>The inner eyebrows are bent upward. The eyes are slightly 
      closed. The mouth is relaxed.</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="7%"><FONT size=1>
      <P align=justify>3</FONT></P></TD>
    <TD vAlign=top width="32%"><FONT size=1>
      <P align=justify>Anger </FONT></P></TD>
    <TD vAlign=top width="60%"><FONT size=1>
      <P align=justify>The inner eyebrows are pulled downward and together. The 
      eyes are wide open. The lips are pressed against each other or opened to 
      expose the teeth. </FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="7%"><FONT size=1>
      <P align=justify>4</FONT></P></TD>
    <TD vAlign=top width="32%"><FONT size=1>
      <P align=justify>Fear </FONT></P></TD>
    <TD vAlign=top width="60%"><FONT size=1>
      <P align=justify>The eyebrows are raised and pulled together. The inner 
      eyebrows are bent upward. The eyes are tense and alert.</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="7%"><FONT size=1>
      <P align=justify>5</FONT></P></TD>
    <TD vAlign=top width="32%"><FONT size=1>
      <P align=justify>Disgust </FONT></P></TD>
    <TD vAlign=top width="60%"><FONT size=1>
      <P align=justify>The eyebrows and eyelids are relaxed. The upper lip is 
      raised and curled, often asymmetrically.</FONT></P></TD></TR>
  <TR>
    <TD vAlign=top width="7%"><FONT size=1>
      <P align=justify>6</FONT></P></TD>
    <TD vAlign=top width="32%"><FONT size=1>
      <P align=justify>surprise </FONT></P></TD>
    <TD vAlign=top width="60%"><FONT size=1>
      <P align=justify>The eyebrows are raised. The upper eyelids are wide open, 
      the lower relaxed. The jaw is opened. </FONT></P></TD></TR></TBODY></TABLE><FONT 
size=1>
<P align=justify>&nbsp;</P></FONT>
<P align=center><IMG height=180 
src="&#65288;7&#65289;Face and 2-D Mesh Animation in MPEG-4.files/Image21.gif" width=378></P>
<P align=center><A name=_Ref417884212>Figure 3</A>: Facial expressions.</P><A 
name=_Ref447380710>Face </A>Model Specification </B>
<P align=justify><B></B>Every MPEG-4 terminal that is able to decode FAP 
streams, has to provide an MPEG-4 compliant face model that it animates (Section 
2.1.3.1). Usually, this is a model proprietary to the decoder. The encoder does 
not know about the look of the face model. Using a FDP (Face Definition 
Parameter) node, MPEG-4 allows the encoder to completely specify the face model 
to animate. This involves defining the static geometry of the face model in its 
neutral state using a scene graph (Section 2.1.3.3), defining the surface 
properties and defining the animation rules using Face Animation Tables (FAT) 
that specify how this model gets deformed by the facial animation parameters 
(Section 2.1.3.4). Alternatively, the FDP node can be used to 慶alibrate