No, dammit, there's a dumb bug. I will post a fix this morning.
Sorry about that.
-- lg
> On Jan 27, 2021, at 5:51 AM, Daniel Flehner Heen <[email protected]>
> wrote:
>
> Hi!
>
> I'm attempting to implement the use of ImageBufAlgo.ociolook(), but keep
> getting an error message "3D Luts can only be applied in the forward
> direction. (inverse) specified." even if I provide "inverse=False"
> oiio.ImageBufAlgo.ociolook(
> buf,
> buf,
> "Twin",
> "acescg",
> "out_rec709",
> inverse=False,
> colorconfig=ocio_path,
> roi=roi
> )
> Am I misunderstanding what the inverse argument is used for?
>
> The look definition in the config.ocio file:
> - !<Look>
> name: Twin
> process_space: ACES - ACES2065-1
> transform: !<GroupTransform>
> children:
> - !<ColorSpaceTransform> {src: ACES - ACES2065-1, dst: AlexaV3LogC}
> - !<FileTransform> {src: twin_FC5213normal_v4.cube, interpolation:
> linear}
> - !<ColorSpaceTransform> {src: Output - Rec.709, dst: ACES - ACES2065-1}
>
> If I provide anything in inverse_transform like:
> inverse_transform: !<ColorSpaceTransform> {src: Output - Rec.709, dst: ACES -
> ACES2065-1}
> to the config, ImageBufAlgo.ociolook() stops complaining, but only applies
> the transform inversed no matter what the inverse argument in ociolook says.
>
> I confirmed this by applying the same look in Nuke and setting it to inverse.
>
> So it seems ociolook() only favors inverse for some reason.
> Does anyone else have any experience with this?
>
> oiio2.2.10.0
> ocio1.1.1
>
> --
>
> -Daniel
> _______________________________________________
> Oiio-dev mailing list
> [email protected]
> http://lists.openimageio.org/listinfo.cgi/oiio-dev-openimageio.org
--
Larry Gritz
[email protected]
_______________________________________________
Oiio-dev mailing list
[email protected]
http://lists.openimageio.org/listinfo.cgi/oiio-dev-openimageio.org
