> 0Root EntrygE S( F +NmC9szWordDocument U1794156NX5CompObjjSummaryInformation( !"#$./123456789:;DEFGHIJKLMNRoot EntrygE S( F +Nm`)3szWordDocument U1794156NX4CompObjjSummaryInformation(
<% !"#$&'()*+,=>?@ABCܥhW De5
ll""""",$*",$,$,$,$H$",$J41j$:$$$$$$$&&&&&P)(Py){4X4A)
"$$$$$)%""$j$%%%$"$"$&C9sz""""""$&%%
.ASSSS 143keyword void should be bold269delete (Fij will also be wavelength dependent, but this is ignored)Fig 11.14ܥhW 4e4
ll"""""$*"$$$$4$"$
41V$:$$$$$$$&&&&&P'PE);4X4Ac)
"$$$$$c)%""$V$%%%$"$"$& Kz""""""$&%%
.ASSSS 143keyword void should be bold269delete (Fij will also be wavelength dependent, but this is ignored)Fig 11.14 ;<=@AHKMl{  } ~
!
8
9
:
=
>
B
J
L
g
q

TUVZ[dxyJr]c]ch]cu]U]cX <=A
!
9
:
>
UV[''''''''''='='''''''''''k +48
) 48
) K@Normal]a "A@"Default Paragraph Fontconfusing top left subcaption: f and g overlap  whether f projects onto one or two pixels would depend on the order in which the patches were processed (given that they are at equal distances). The diagram could be changed so that the 2 extent above f was changed to 1.
P281 sentence beginning Recall, with reference to the figure...could be deleted  all p0olygons in the figure are on the same plane.274first (i should be (j271Eqn 11.4 n should be N348the sentence After building this table subsequent arclength calculations are greatly decreased should be ...the cost of subsequent arclength calculations is greatly decreased443there are two WILL83 references. The first can be deleted  it is not referenced in the text345 reference [THAL85] should be [MAGN85]362Eqn at top of RH col. Should read W = (trace) EMBED Equation.2
3rd line should read: the sine of half the angle of rotation...371Eqn at bottom of LH col. tan EMBED Equation.2 missing: should be (1 = tan EMBED Equation.2 (RHS).ASSSS I am indebted to Prof. Nelson Max (UC Davis) for the following:
gure... could be deleted  all p as far as I can see
then [WILL83b] on page 189 becomes [WILL83]373link notation wrong: LH link should be i23794th
234:EFGHI]^_befgmxyz{uD:]ceKuD:]avK]chJq]cuDf:]ceKuDf:]avK]chuD,:]ceKuD,:]avK]cuD]cBgtu(+5TpsBCFNP_`aKMX`u]ch]ch]c8DocumentSummaryInformation8
ObjectPooly4zy4z_984296645F@4zNm՜z_984296806F0tzNm՜zws 95@VP@03m@ҧ#z@Љ]z՜.+,0HPpxLightwork Design LtdMeeting 9th OctoberܥhW elllll
FMicrosoft Word Document
MSWordDocWord.Document.69qOh+'0 $0
Xdp
Meeting 9th OctoberAdministratorANormal.dotAdministratorl6SMicrosoft Word for Windo_984296880
F`zNm՜zOle
TPIC
RLMETAK
!
@AEbfNRim?CX?Cn!&dhLT
'''''''=''''''''P''='''''''''''g(
@TV
:::or
mpGJNQ&)UZ]b2 6 P Z A
G
v
{
Administrator%E:\Users\Alan\Meeting 9th October.doc
AdministratorE:\Users\Alan\AART_erratall
1 3 P
P"XzAlll
ll wzllll b
.ASSSS u]U]c''48
) K@Normal]a "A@"Default Paragraph Font b
Administrator%E:\Users\Alan\Meeting 9th October.doc
AdministratorE:\Users\Alan\AART_errata.doc@rlab1Ne02:winspoolrlab1rlab1G 4d,A4PRIV''''trlab1G 4d,A4sPRIV''''t1Times New RomanSymbol&Arial"hg3Fg3F)f$AMeeting 9th October
Administrator
AdministratorCompObj
IfEquation Native H4ObjInfoEOle
aPIC
\LMETAUEquation Native `4CompObj^fObjInfoFOle
nPIC
iLMETAbEquation Native m=CompObjkfObjInfoGl
1
FMicrosoft Equation 2.0DS EquationEquation.29q .1`&P&MathType Symbol
b~wk~wWw
 2
>{ Times New Romank~wWwN
 2
1p
&
"System
L, .1`&P&MathType Symbol
b~wk~wWw
 2
>{ Times New Romank~wWwN
 2
1p
&
"System
L,
FMicrosoft Equation 2.0DS EquationEquation.29ql
1=N .1 &&MathType `@` Times New Romank~wWwU
 2
O1p 2
LK2p
&
"System
L=N
FMicrosoft Equation 2.0DS EquationEquation.29q!l
12.doc
AdministratorE:\Users\Alan\AART_errata.doc
AdministratorE:\Users\Alan\AART_errata.doc
AdministratorE:\Users\Alan\AART_errata.doc@rlab1Ne02:winspoolrlab1rlab1G 4d,A4PRIV''''trlab1G 4d,A4sPRIV''''t
c
@ADE`abefmr LMNQRYghilm'<=>?BCXbd
()/89:@ABHSTUVWY[agn
=>?BCDFr !%&bcdghz{`f IJKLST8 9 ; ?
u
v
{
0!
0101 0;0<1=0@1A1H1M00101l0{ 1 1} 1~ 1a
1 1 1 1b
1i
1k
1 0
0
1
0
1
1
0
0
1
0
1!
08
09
1:
0=
1>
1B
1J
1L
1g
1q
1
1
1
1
1
1
1
1
1
0
0
1
0
1
1111
0
0T0U1V0Z1[1d1x1y100101111111111111110111111
001011213141:1E1F1G1H1I1b1e1f1g1m1x1y1z1{1001
01
1
001
01
1
01,001V01Y1a1b1g0010111110010100101001(01+151T001p01s1111111111B001C01F111001011N1P1_1`1a1111111001K01M1X1`0001Times New RomanSymbol&Arial"qhg3FV3F)f$AMeeting 9th October
Administrator
Administrator1B
1J
1L
1g
1q
1
1
1
1
1
1
1
1
1
0
0
1
0
1
1110T0U1V0Z1[1d1x1y10010111 ;<=@AHKMl{  } ~
!
8
9
:
=
>
B
J
L
g
q

TUVZ[dxyJr]c]ch]cu]U]cX <=A
!
9
:
>
UV[''''''''''='='''''''''''k +48
) 48
) K@Normal]a "A@"Default Paragraph Fontconfusing top left subcaption: f and g overlap  whether f projects onto one or two pixels would depend on the order in which the patches were processed (given that they are at equal distances). The diagram could be changed so that the 2 extent above f was changed to 1.
P281 sentence beginning Recall, with reference to the figure...could be deleted  all p0olygons in the figure are on the same plane.
JLMNOPQSVWXYZ[]_cdefghjl274first (i should be (j271Eqn 11.4 n should be N348the sentence After building this table subsequent arclength calculations are greatly decreased should be ...the cost of subsequent arclength calculations is greatly decreased443there are two WILL83 references. The first can be deleted  it is not referenced in the text345 reference [THAL85] should be [MAGN85]362Eqn at top of RH col. Should read W = (trace) EMBED Equation.2
3rd line should read: the sine of half the angle of rotation...371Eqn at bottom of LH col. tan EMBED Equation.2 missing: should be (1 = tan EMBED Equation.2 (RHS).ASSSS I am indebted to Prof. Nelson Max (UC Davis) for the following:
gure... could be deleted  all p as far as I can see
then [WILL83b] on page 189 becomes [WILL83]373link notation wrong: LH link should be i23794th
234:EFGHI]^_befgmxyz{uD:]ceKuD:]avK]chJq]cuDf:]ceKuDf:]avK]chuD,:]ceKuD,:]avK]cuD]cB!
a
b
i
k
!"(,LMOPRSVYabg]chJq]c]chU]cu]cZ'P'P'''''''''''''''''''''''k +48
) %'''''''''''''''''''''''''48
) k +%''''''''''''''''''''''''k +48
) %a
('''''''''''=''48
) k +48
) Eqn RH col: RHS should read r + (n x r
LHS subscript notation should be (rx,ry,rz,1)380LH col. First 3 eqns: all Pj should be Pi390first para Sect 16.6.1 walking would have been better than legged motion407 fsw curve should be smooth  that is no corner discontinuity419figure upside down433[FOUR86] reference is missing the coauthor of this paper W.T. Reeves425LH col: The sentence beginning In this case... is confusing/wrong. Add full stop after z=0 and delete the remainder of the sentence. Replace with This summation will exhibit a maximum and minimum value.410captions for (b) and (c) angles do not correspond to figure. All captions (a) (b) and (c) can be deleted since they do not add anything to the point of the illustration.401402(u,v,w)/(s,t,u) notation: u has been used in two different contexts. To the end of the 2nd para. LH col. P 401 it has been used as the first parameter in a tricubic Bezier patch definition. From the beginning of the next para. It is used as the third coordinate in lattice space. This confusion is compounded on p 402 where both uses appear on each side of the equation at the bottom of the first column. I think this is a case where it is better that the reader realises this rather than renotate. To change means altering all occurrences of (s,t,u) to (r,s,t) and similarly (S,T,U) to (R,S,T).38first eqn: integral sign should follow equals sign.ASSSS Ne02:winspoolrlab1rlab1G 4d,A4!
a
b
i
k
!"(,LMOPRSVYabg]chJq]c]chU]cu]cZ'P'P'''''''''''''''''''''''k +48
) %'''''''''''''''''''''''''48
) k +%''''''''''''''''''''''''k +48
) %a
('''''''''''=48
) k +48
) Eqn RH col: RHS should read r + (n x r
LHS subscript notation should be (rx,ry,rz,1)380LH col. First 3 eqns: all Pj should be Pi390first para Sect 16.6.1 walking would have been better than legged motion407 fsw curve should be smooth  that is no corner discontinuity419figure upside down433[FOUR86] reference is missing the coauthor of this paper W.T. Reeves425LH col: The sentence beginning In this case... is confusing/wrong. Add full stop after z=0 and delete the remainder of the sentence. Replace with This summation will exhibit a maximum and minimum value.410captions for (b) and (c) angles do not correspond to figure. All captions (a) (b) and (c) can be deleted since they do not add anything to the point of the illustration.401402(u,v,w)/(s,t,u) notation: u has been used in two different contexts. To the end of the 2nd para. LH col. P 401 it has been used as the first parameter in a tricubic Bezier patch definition. From the beginning of the next para. It is used as the third coordinate in lattice space. This confusion is compounded on p 402 where both uses appear on each side of the equation at the bottom of the first column. I think this is a case where it is better that the reader realises this rather than renotate. To change means altering all occurrences of (s,t,u) to (r,s,t) and similarly (S,T,U) to (R,S,T).38first eqn: integral sign should follow equals sign.ASSSS b1Ne02:winspoolrlab1rlab1G 4d,A4gtu(+5TpsBCFNP_`aKMX`u]ch]ch]c9
!
@AEbfNRim?CX?Cn!&dhLT
'''''''=''''''''P''=''''''''''''g
@TV
:::ormpGJNQ&)UZ]b2 6 P Z A
G
v
{
Administrator%E:\Users\Alan\Meeting 9th October.doc
AdministratorE:\Users\Alan\AART_errata.doc
AdministratorE:\Users\Alan\AART_errata.doc
AdministratorE:\Users\Alan\AART_errata.doc
AdministratorE:\Users\Alan\AART_errata.doc
AdministratorE:\Users\Alan\AART_errata.doc@rlab1Ne02:winspoolrlab1rlab1G 4d,A4PRIV''''trlab1G 4d,A4sPRIV''''t
c
@ADE`abefmr LMNQRYghilm'<=>?BCXbd
()/89:@ABHSTUVWY[agn
=>?BCDFr !%&bcdghz{`f IJKLST8 9 ; ?
u
v
{
0!
0101 0;0<1=0@1A1H1M00101l0{ 1 1} 1~ 1a
1 1 1 1b
1i
1k
1 0
0
1
0
1
1
0
0
1
0
1!
08
09
1:
0=
1>
1B
1J
1L
1g
1q
1
1
1
1
1
1
1
1
1
0
0
1
0
1
1111
0
0T0U1V0Z1[1d1x1y100101111111111111110111111
001011213141:1E1F1G1H1I1b1e1f1g1m1x1y1z1{1001
01
1
001
01
1
01,001V01Y1a1b1g0010111110010100101001(01+151T001p01s1111111111B001C01F111001011N1P1_1`1a1111111001K01M1X1`00001Times New RomanSymbol&Arial"qhg3Fl3F)f$AMeeting 9th October
Administrator
Administrator