Sticky

P: Introducing the Project Indigo camera app

Forum|Forum|10 months ago
May 23, 2025
394 replies
240919 views

This post applies to the Project Indigo iOS camera app.

Adobe Labs is excited to share an early look at Project Indigo, an iPhone camera app we've started to develop, to get feedback from the photography community. The app offers full manual controls, a more natural ("SLR-like") look, and high image quality in both JPEG and raw formats. It also introduces some new photographic experiences not available in other camera apps. For more information on the underlying technology, please refer to this Project Indigo blog post.

Before you start with Project Indigo

We recommend using Project Indigo on iPhone 15 Pro/Pro Max or newer devices.
(Also supported are 12 Pro/Pro Max, 13 Pro/Pro Max, and all 14-series devices.)

You should have at least 1GB of storage space left for the app, the downloadable AI Models inside the app, and for captured photos.

Download the app from the App Store.
Check out the Project Indigo FAQ, which is updated regularly with new explanations.

Recipes for success when using Project Indigo

To get the maximum out of your images captured with the app, follow these guidelines:

When reviewing the results, focus on Project Indigo's more natural look (in both SDR and HDR). If you haven’t done this before, try viewing the images on your laptop or desktop device, preferably on an HDR screen.

Capture with both JPEG and raw DNGs with file saving enabled. Project Indigo produces computational photography DNG files, which have the same natural look as JPEG images, but much more latitude for editing after capture.

Take control of the camera with the built-in Pro Controls, including controls that are exclusive to a computational camera: Frames to Merge and Merge Method. These may be intimidating for beginners, but with Project Indigo, you can try them for free, and nothing will break—you can always reset the settings to ‘Auto’ and let the camera take back control.

Go to the Indigo Labs page and play with the latest innovations our team can offer. These are only available on mobile via Indigo!

Be patient! Project Indigo is doing a lot of heavy lifting under the hood, and it will reward you with great photos. In return, it may ask you for a bit of time to set up captures when needed, and to wait a few seconds for the image processing to finish.

Sending feedback

Please try the app and share feedback in this community forum thread. If you report a problem you encountered, it would help to include details like which device you are running Project Indigo on, what kind of scene you were trying to capture, what you were trying to achieve with the camera, and as much information as possible about what you like or do not like about the resulting photo quality. Our team will continually monitor this thread to track issues and improve future experiences.

To improve the performance and results of Project Indigo, it is important that examples of images that do not meet your expectations are forwarded to the team via your report. A large variety of file formats are allowed as attachments in these forum posts. The best option is to attach your image's raw file directly to your feedback post. Note that there is a 50 MB limit on an attachment's file size. If your raw file is too large to attach, the best option is to share the file via a file-sharing service (Dropbox or similar) and then share the link in your feedback post. Thank you for continuing to provide feedback on the Project Indigo camera!

Boris Ajdin: Product Manager, NextCam

Posted by:

J

Jeff Donald

Participating Frequently

I’ve grown weary of reading the posts of nhan_8084 never ending stream of gibberish. I thought my experiences would possibly help in some small way, but this forum/community lacks any kind of meaningful moderation. I’ve found a much better managed community that welcomes participation from all members. I wish everyone the best with the stream of consciousness posts.

B

BorisTheBlade

Adobe Employee

Hi @Jeff Donald - thank you for your input. The lack of moderation is solely on me, as I am trying my best not to be a “dictator” of any kind and let the users passionate about Indigo have space to voice their opinions. Having said that, I should help everyone be a bit more structured. So let me try to do that. I would find your feedback very valuable, so please, do not quit on the forum and post your thoughts when you see fit! Thank you for trying to help make Indigo better.

N

nhan_8084

Known Participant

@BorisTheBlade found a issue where tele shot gave a odd greenish blue tint! This is day light condition, both shot auto as usual, 4x and 8x.

There’s also a very heavy blue tint from a indoor shot using the tele as well! This shot is 4x of the toddler with censored face for privacy purposes. Lighting is behind me, left of him from a window, window is a dual window so it’s quite wide for bright light. Stock wb and expo is always spot on as mentioned, so I have attached that as well.

I also attached two crops of hairs. You can see that the crop that has fine grains, that’s the style and noise reduction (less) that I like. That shot is the tele crop. The other crop of hair is from the main sensor, much larger sensor than the tele, brighter aperture lens as well, so it SHOULD NOT need to have that much denoising applied! The hair is all smoothed out, the walls and back ground has no luma grains either.

B

BorisTheBlade

Adobe Employee

Thank you for all your tests. It takes time to evaluate everything and sometimes to even understand what exactly you are trying to point out as a flaw in Indigo, so please bare with us. On a suggestion from Jeff above, I would like to ask you to consider creating one thread where you’ll keep adding more experiments, instead of creating a new main forum message every time? This way the main forum may not be “overwhelmed” by the volume of your inputs, while you’d still get to share all your findings. Plus, it may be easier for me to find what I am looking for if it is always in the same thread. I’d suggest adding one thread per release for image quality-related questions, and one more (if you feel like it) for workflow issues, bugs, miscellaneous, etc. Thank you!

N

nhan_8084

Known Participant

Hello @BorisTheBlade , every single thread I created is of findings that is not related to the other. They al have different issues at time of shooting, and if you read at every single one of them it clearly explains what the issues are. There’s not a post that I made that isn’t clear. If they aren’t let me know, as I am not a fan of using circles or arrows to show the areas of attention because I know you and your team have vast knowledge to pinpoint the issues if a versus is shown.

None of my posts are irrelevant or spams posts as I said because I show issue as they arises, and report them as you said to do, how else would I show them? Maybe the users here do not care on the pixel peeping level like I do, but how can you point that out as my problem or as Jeff said gibberish? When you are “spoiled” by gcam and the output that it does, not on one device, but multiple devices of android and you see iOS with nothing to compare to, it gets weary just seeing how bad iPhone pictures are.

I did not also see anyone posting issues and show detailed analysis like I do so perhaps they really don’t care? I don’t bury people’s posts either so Jeff attacking me is unfair and not tolerated no?

However, if you want me to post as you said one post I can do that and subpost in that sure.

N

nhan_8084

Known Participant

@BorisTheBlade I just realized that Indigo has two different tonemapping going on for jpeg and raw? The SOOC jpeg has its own looks while the raw when imported to lrm and export to jpeg without touching anything has it's own look!

Dng to jpeg nothing touch, just hit export: https://www.dropbox.com/scl/fi/eiwdqa0go2rc01r3jsp01/IMG_4454.JPG?rlkey=8ibdmag4wq38srplt8lh2dw2y&st=o0hc1rw0&dl=0

Shouldn't the raw when imported has the same look as the jpeg bc it is afterall processed and tonemapped, why would raw has a dif look and jpeg has a dif look. If anything, I rather have the SOOC have the look that the raw showed when imported to lrm (but have no sharpening , less denoising, luma and chroma). As you can see from the sample, the potbelly building is brighter, skies are not crushed blacks!

N

nhan_8084

Known Participant

@BorisTheBlade it seems like non-SR shots (ie.. optical range 0.5x, 1x, 4x) has the EXACT same artifacts as SR shot when there’s a slight motion detected in frame while capturing. The samples attached (cropped) showed a moving toddler (not fast motion at all, just normal speed human movements). I cannot show the full image due to privacy concern ofc, but the stock app do not have any artifacts at all and I will show the crops too!

What is even more strange is that the Indigo SOOC jpeg has artifacts in areas of the frame that doesn’t contain movements, like the water bottle. I mentioned before that I encountered this same artifacts in gcam and fixed it by using a custom generated noise model four channels (A through D).

You can obviously know which one is the stock app so I do not need to spell out which pictures belong to which. The exif is also included. By the way, the raw of Indigo also have it of course bc the JPEG is generated from that raw

PLEASE PLEASE test motion on your end using the 17 pro max, normal indoor lighting where ss drops to about 1/60s to 1/80s as shown.

N

nhan_8084

Known Participant

I've got another scene where it's heavily under exposured, matching your statement earlier of a bright source involved that can do that, it's similar scene you have seen before and that's with my grow light. To remind you the grow light temp is 4000K per the vendor.

All shot auto, using main lens. Indigo raw is attached, SOOC jpeg, proraw embedded 12mp jpeg, processed indigo raw to jpeg to represent true to scene, screen shots of how I did it (quite a lot sadly), and screenshot comparing issues of SOOC jpeg to the post processed. My version of jpeg will always noise reduction, and sharpness killed and you can see that it still have more details than the AI sharpening. I believe as I say all along, denoising needs to be completely set to 0, including both chroma and luma, or it will be not pleasing at all. The processed jpeg has more noise bc it's obviously denoise set to 0, but also because you have to raise shadow and expo while keeping hdr high lights vs it being clipped in the SOOC jpeg. Very challenging situation to have a balance, but I chose to leave nr to 0.

DNG Post Processed: https://www.dropbox.com/scl/fi/6n1u3xd5lq0wpi4cx68us/IMG_4342.JPG?rlkey=4zoluimw14vcns8dqhl95bbr1&st=uvf6chf5&dl=0

IMG_4340: SOOC Indigo jpeg

IMG_4341: 12mp ProRAW embedded jpeg (ignore the .dng dunno why it only share the jpeg)

Rest are screenshots showing the obvious: lrm settings used, and issues where SOOC jpeg are (crushed shadows, lack of details, mush, highlights clipped).

I really hope the team take my findings serious and rework these existing issues, as I work extensively and a lot of time spent taking shots and analyzing them.

Simply put, put the team's views and your views in the shoe of the users like me. Would you want to use and share these images I showed here as is to friends and families, colleagues, etc... or share the properly rendered jpeg that is true to scene and capturing it without having to so so much adjustments?

Photo Apr 11 2026, 11 16 39 PM.dng

IMG_4341.DNG

N

nhan_8084

Known Participant

Here's a scene taken in the late evening, 6:30 pm area est time zone. As you can see the Indigo SOOC jpeg is definitely vibrant and punchy, looks good at first as is, until you zoom in 100% crop or more. Colors not only isn't realistic, there's heavy black crushing shadows while in reality shadows aren't black.

I believe I found the "source" for the blobs looking, or mush clumps of foliages, and that is Indigo AI has too much raw sharpen, and raw contrast. I've included the raw, and post processed to jpeg to look true to scene. You will see that Indigo is still sharp, with sharpening to 0, noise reduction and color nr to 0 in lightroom mobile. I also include crops to show where the post process excels vs the SOOC jpeg. There's not a need for the team to have THIS MUCH digital sharpening, or color denoising since the post processed jpeg is amazing already at detail and very minimal noises with all nr to 0!

As usual, everything auto. Please remind the forum team to allow jpeg and png (screenshot) to have unlimited size upload so I don't keep using dropbox for files bigger than 5mb!!

Raw post processed: https://www.dropbox.com/scl/fi/n130lvtxcxb6wys25wdyd/IMG_4332.JPG?rlkey=nk2ow4iujp3i1sdsptlrojcsj&st=67o7foy7&dl=0

Indigo SOOC jpeg: https://www.dropbox.com/scl/fi/tbfnwrytnb0ozk7sdwn6b/IMG_4326.JPG?rlkey=2w5ka0vpx5bi8laczpa8ar3aw&st=zeyj1ebw&dl=0

Photo Apr 11 2026, 6 22 03 PM.dng

N

nhan_8084

Known Participant

Ok now let's focus on textures, noise to details rendition. The following sample is of the backyard scene but is a comparison with the app Camac. All is shot in auto, no tapping, nothing changed on either app.

Camac jpeg: https://www.dropbox.com/scl/fi/x96lrkphsqy8ish45ynhl/IMG_4292.JPG?rlkey=qnojej80m0utqws39a9uvcgnl&st=ve722h4i&dl=0

As you can see from the compare app, look at how much clumpy foliages are comparing to Camac? Again, I remind you this is untouched jpeg out the cam for both. Camac is not oversharpened either, it's just AI still denoising way too much. Look into the radius denoising perhaps? I do not know how you guys call your parameters, but radius denoising is what I am used to in Gcam.

Also, Indigo has crushed shadows due to the heavy tone mapping done. Indigo might look pleasing and social media ready since all social media compresses image to eternity, but it is definitely not usable or pleasing if that photo was to be printed.

All I want to ask of Indigo's team is to keep things as reality, as how the human eyes sees things, not whether the team likes the looks or not bc that will now become subjective, vs the objective: to look like the true to scene!

N

nhan_8084

Known Participant

@BorisTheBlade here are two samples showing extreme over-saturations and incorrect colors of red(burgundy ish). Red is an extremely difficult color to get in photography let alone allowing AI to do the work. I will attach the as is jpeg, how it looks as soon as you import to lrm, and after you Reset to All. The dng will also be included. The Reset to All is the correct color btw!

Import dng as is: https://www.dropbox.com/scl/fi/efx9abaibuma8m3392aay/IMG_4302.png?rlkey=olqimm1da73rdk4ozb06h9e8t&st=n7jqirbt&dl=0

Reset to All: https://www.dropbox.com/scl/fi/g7oj8akput1rqmeyz8vok/IMG_4303.png?rlkey=ygs118l52923n8xy138xsf2go&st=gp7no8q2&dl=0

Pay attention to the Rose color as I said, the hues and tones are massively different.

The other sample is showing extreme oversaturation, like if you have vivid profile, or a extreme HDR pop profile, not natural at all. Anyone looking at the JPEG as is will be disgusted at how it looks vs reality. I disagree with you when you mention saturations and exposure is subjective, as the goal of capturing ANY photos is to capture the reality as how you see it… call it Kodak Moment if you will.

Only Colors sliders changed via LRM: https://www.dropbox.com/scl/fi/wwj714wp8mz8iirh36izc/IMG_4298.png?rlkey=32y9k7b2v6jkno89ywe632mrf&st=tunls0jq&dl=0

Exported to JPEG: https://www.dropbox.com/scl/fi/df1mucmg4e17fiyr7dmpo/IMG_4299.jpeg?rlkey=9n4mwvjkw6zzr2dcj6l6s7hlx&st=ivfr0t86&dl=0

Raw is also included of the backyard scene.

Known Participant

Hello @BorisTheBlade since I made this topic regarding incorrect red hues and color of the rose. This shot is of the same rose tree/bush, just dif angle and lighting. This time the AI rendition of red is much more accurate to real life. However, upon a crop you can see oversharpening digital artifacts, and uneven shades of reds that looks like a gamma issue, color clipping? I don’t exactly know the correct term, but you can tell there are splotching everywhere! You can crop into the jpeg yourself and you will see it looking worse than the attached crop. A member I know who is used to the photography scene call it “notorious 6”.

N

nhan_8084

Known Participant

This is a example of exposure gone wrong using the tele. The wooden object on the wall is near invisible which is NOT how it is in real life. While stock is brighter than reality but not by much. Even the night mode in Indigo cannot recover the proper exposure, and stock night mode adds in more denoising but exposure is similar and consistent.