Notices

I LOVE IT.

just a suggestion.


an on-screen capture button would be awesome.


it would avoid the shaking of the camera whenever we press the physical capture button ergo blurry images

is it me or the latest app and driver are the same thing?

This new driver somewhat makes the HDRcapture application became really terrible at taking picture quality. Jagged edges and noisies.

Is vignetting effect removed completely? Can there be vignetting effect made as an option on the picture?

I want this effect back as an option in HDRcapture please.



This is good if we have an option to use it to take these kind of picture. Looks so pleasant

maxximuscool -

Thanks for noticing the project page mixup. Will fix shortly.

We haven't changed anything about the vignetting in the latest fcam-drivers or fcamera, so we're a bit confused about what you're seeing. It should be just as bad as it's always been with fcam. Similarly, the minor bug fixes we put into the latest fcam-drivers really shouldn't impact hdrcapture at all - it was a fix to a half-resolution mode no public application is using, to our knowledge. However, we at Stanford don't have source for hdrcapture, so I can't say anything about it for sure.

geohsia -

You should use the -u flag with fcamDngUtil:

> fcamDngUtil -u input.dng output.dng

to update a DNG to the latest FCam metadata.

Thanks. It works great. I need to ask, would it be possible to output RAW after Nokia has done its dark frame subtraction and etc, before it outputs to JPEG and before it applies its own tone curve? While I can handle vignette, I noticed that there is significant color variation in the vignette and that is hard to fix in post processing.

I have profiled my sensor using a X-rite Colorchecker Passport but getting the colors to look right in the shadows is extremely difficult. No matter how I manipulate RAW in PS or LR I can't get it to match the even color I get from the stock camera.

I prefer RAW, but right now I am unable to improve on the current camera output (aside from fine detail at 100%). Any suggestions? Thanks.

Unfortunately, the OMAP3 processing pipeline gives you two options:

1) RAW data from the sensor. That's what we put in the DNGs. No processing beyond what the sensor itself does (which amounts to just some analog and digital gain, and for lower resolutions, averaging pixel blocks together).

2) 8-bit UYVY data that's been demosaiced, denoised, color-converted, color-subsampled, and (possibly) resized. This is what we use for the viewfinder.

I don't actually think the built-in camera app does dark frame subtraction in hardware in any case - the OMAP3 imaging pipeline can do it, but only for 8-bit images and doing so disables vignetting compensation, which I believe they do instead. Dark frame subtraction in software is possible for both the built-in app, and for ours - we just haven't implemented it in FCamera.

It's been on our list of things to do to deal with the vignetting issues better, since we currently do nothing about it. But unfortunately it's not looking like it'll happen soon - it's not conceptually that difficult, but requires a lot of testing and experimentation to calibrate properly, which (as you've noticed) takes some time to do.

[QUOTE=ETalvala;833440I don't actually think the built-in camera app does dark frame subtraction in hardware in any case - the OMAP3 imaging pipeline can do it, but only for 8-bit images and doing so disables vignetting compensation, which I believe they do instead. Dark frame subtraction in software is possible for both the built-in app, and for ours - we just haven't implemented it in FCamera.

It's been on our list of things to do to deal with the vignetting issues better, since we currently do nothing about it. But unfortunately it's not looking like it'll happen soon - it's not conceptually that difficult, but requires a lot of testing and experimentation to calibrate properly, which (as you've noticed) takes some time to do.[/QUOTE]

Interesting. Thanks for filling in the details. I'll fully admit that I don't really understand what the camera is doing or is capable of so I appreciate you filling us in.

I guess here's my question. I don't mind doing a fair bit of post processing to the final image to get accurate color. Until you're able to do it in camera, how would you suggest we post process so that we get shadows that are grey vs blue/ green. Have you been able to achieve even color throughout the frame via Photoshop or Lightroom?