(-gptr- pin):
(-int-  background indimi indimj inmodi inmodj ppi ppj):
RETURNS: (-float-)
SIDE EFFECTS: None
AUTHOR: Y. LeCun
COMPILABLE: Yes
DESCRIPTION:
returns a bilinearly interpolated pixel value for coordinate
<ppi> <ppj>. The image data is pointed to by <pin>, with 
<indimi> <indimj> <inmodi> <inmodj> being the dimensions and modulos.
This function clips automatically if <ppi> <ppj> are outside of the
bounds by assuming the outside values are equal to <background>.
pixel values are floats, while coordinates are 32 bit fixed point
with 16 bit integer part and 16 bit fractional part.
The function does not use floating point arithmetics.

((-float-) background):
((-idx2- (-float-)) in out):
((-idx2- (-int-)) pi pj):
RETURNS: Null
SIDE EFFECTS: <out> 
AUTHOR: Y. LeCun
COMPILABLE: Yes
DESCRIPTION:
Warps an image using <pi> and <pj> as tabulated coordinate transforms.
<in> and <out> are idx2 of floats. <background> is the value assumed outside
of the input image. <pi> and <pj> are tabulated coordinates which can
be filled up using compute-bilin-transform or similar functions.
Pixel values are antialiased using bilinear interpolation.

((-int-) background):
((-idx2- (-float-)) in out):
((-idx2- (-int-)) pi pj):
RETURNS: Null
SIDE EFFECTS: <out>
AUTHOR: Y. LeCun
COMPILABLE: Yes
DESCRIPTION:
Warps an image using <pi> and <pj> as tabulated coordinate transforms.
<in> and <out> are idx2 of floats. <background> is the value assumed outside
of the input image. <pi> and <pj> are tabulated coordinates which can
be filled up using compute-bilin-transform or similar functions.
This is essentially identical to warp-fimage, except no antialiasing
is performed (it goes a lot faster, but is not nearly as nice).

((-flt-) x1 y1 x2 y2 x3 y3 x4 y4 p1 q1 p3 q3):
((-float-) background)
((-int-) mode):
((-idx2- (-float-)) in out):
RETURNS: Null
SIDE EFFECTS: <out>
AUTHOR: Y. LeCun
COMPILABLE: Yes
DESCRIPTION:
transforms fimage <in> (idx2 of float) mapping quadrilateral <x1> <y1> <x2>
<y2> <x3> <y3> <x4> (points are numbered clockwise starting from upper left)
to rectangle whose upper left and lower right points are <p1><q1>, <p3><q3>.
result is put in <out> (idx2 of float).  Clipping is automatically performed.
pixels outside of the bounds of <in> are assumed to have the value
<background>.  <mode>=0: no antialiasing, <mode>=1: antialiasing with
bilinear interpolation (2 times slower).  execution time on sparc 10 is about
5 ms in mode 0 and 10 ms in mode 1 for a 32x32 target image.