Not to be left out...

My test with HDRI: just a simple sphere placed on my dining room table. I
made the hdr image with a reflective garden ball, my digital camera and
hdrshop. The shadow is "fake", rendered using MLPov projection trick, with a
little blurring and transparancy adjustment in photoshop. The shadow could
have just as well been created with photoshop alone, but I wanted to test
the projection method for future application with more complex shadows. All
figuring out included (and several tweaking renders) total creation time was
about 2 hours including the time to make the hdr map.

Anyway, on to more complex objects, but wanted to share my little test.

Post a reply to this message

Attachments:
Download 'hdri composite.jpg' (118 KB)

Preview of image 'hdri composite.jpg'

I like this! A good, practical application of HDRI. I think I would have
chosen an object more likely to really be on that table, but heck, for all I
know you have a spherical moonstone sitting on your table in real life. ;)

I think what really gives it away is the lack of jpeg artifacts on the ball.
If you could compress it to the same level as your photo, I think you'd
really have a convincing image.

Hm, I wonder how hard it would be to write a Photoshop filter to simulate
jpeg artifacting? Worth researching a little, anyway.

Good image!

gilrain

Post a reply to this message

Now we're talking...  I think this is the first image I've seen that
actually uses HDRI as it was intended.  Very nice results.  I look forward
to seeing more complex scenes done in a similar manner.  Keep us posted...

Question to others:  How would the IRTC community look upon HDRI scenes
composited into photographed surroundings?  Since this is more than just
using an image map for the background...

Post a reply to this message

Jack,

What digital camera did you use? Mine does not allow exposure
control so it can't be used for the multiple exposure levels needed
for a HDRI. I wonder if a regular film camera would work? If I
could use a 35mm camera the resolution would be higher also, so
I wonder why they typically use a digital. Maybe it is easier, quicker,
no processing and scanning, and they happen to have very good
digital cameras.

HB

"Jack Couilliard" <jack underscore couilliard at hotmail dot com> wrote in
message news:3e20fd18@news.povray.org...
> Not to be left out...
>
> My test with HDRI: just a simple sphere placed on my dining room table. I
> made the hdr image with a reflective garden ball, my digital camera and
> hdrshop. The shadow is "fake", rendered using MLPov projection trick, with
a
> little blurring and transparancy adjustment in photoshop. The shadow could
> have just as well been created with photoshop alone, but I wanted to test
> the projection method for future application with more complex shadows.
All
> figuring out included (and several tweaking renders) total creation time
was
> about 2 hours including the time to make the hdr map.
>
> Anyway, on to more complex objects, but wanted to share my little test.
>
>
>

Post a reply to this message

> What digital camera did you use?

Olympus c-3040 (3 Megapixel). Not the world's best digicam, but certainly
not the worst. I held my f-stop at 2.8 and shot 12 shots in tiff mode with
auto whitebalancing turned off. Shots were spaced one exposure apart using
different shutter speeds (ie 1/800, 1/400, 1/200, etc). I also used a remote
control and tripod to minimize pixel drift from one shot to the next. 12
shots was probably overkill, I am planning on using 5 or so for some tests
in the future.

> I wonder if a regular film camera would work?

I'm sure it would, but you have the extra step of scanning in your negatives
(I don't think you would be able to get good results scanning prints, which
only hold 200dpi or so). Scanning 35mm file might even have a couple
advantages over digital. Film would provide a constant "white balance",
although in many cameras you can turn auto whitebalancing off. Also, good
negative scanners can capture a higher dynamic range (10, 12 or 16
bits/pixel) than digital cameras, decreasing the number of braketed shots
needed to make an HDRI image.

Anyway, good luck. I hope to see some of your results.

Jack

Post a reply to this message

Jack,
Thanks for the info, I will try using film and scanning.
I was wondering why you used an f-stop of 2.8, I would
think that maximum depth of field is desired, so an f-stop
of 16 or 22 would be better. Stopping down, you would
then use longer shutter speeds but get better
sharpness across the depth range. I think.

Harold


"Jack Couilliard" <jack underscore couilliard at hotmail dot com> wrote in
message news:3e24a247$1@news.povray.org...
> > What digital camera did you use?
>
> Olympus c-3040 (3 Megapixel). Not the world's best digicam, but certainly
> not the worst. I held my f-stop at 2.8 and shot 12 shots in tiff mode with
> auto whitebalancing turned off. Shots were spaced one exposure apart using
> different shutter speeds (ie 1/800, 1/400, 1/200, etc). I also used a
remote
> control and tripod to minimize pixel drift from one shot to the next. 12
> shots was probably overkill, I am planning on using 5 or so for some tests
> in the future.
>
> > I wonder if a regular film camera would work?
>
> I'm sure it would, but you have the extra step of scanning in your
negatives
> (I don't think you would be able to get good results scanning prints,
which
> only hold 200dpi or so). Scanning 35mm file might even have a couple
> advantages over digital. Film would provide a constant "white balance",
> although in many cameras you can turn auto whitebalancing off. Also, good
> negative scanners can capture a higher dynamic range (10, 12 or 16
> bits/pixel) than digital cameras, decreasing the number of braketed shots
> needed to make an HDRI image.
>
> Anyway, good luck. I hope to see some of your results.
>
> Jack
>
>

Post a reply to this message