Album AI images from Dave C

Runing hare

nobody@ipernity.com (Dave C) — Thu, 16 Jul 2026 08:59:46 +0000

Dave C has posted:

A while ago I spent a couple of weeks playing around with ai images. This was one of the last attempts I made at trying to produce a realistic natural image.
The ai was given one of my own photographs for reference.

Best viewed full screen.

Thank you for looking and for your kind comments.

The Warrior Pixie

nobody@ipernity.com (Dave C) — Fri, 29 May 2026 06:56:30 +0000

Dave C has posted:

The image looks good at first glance. Just don't start counting legs or antennae on the hornet or the extra thumb on the pixie warrior. ;-D

Thank you for looking and for your kind comments.

Obsolescence

nobody@ipernity.com (Dave C) — Mon, 25 May 2026 11:31:03 +0000

Dave C has posted:

“The story”
(The story of the old Gentleman is based upon a true event)

An elderly gentleman is in the doctors surgery to book an appointment. the receptionist tells him that he can no longer make an appointment face to face with the receptionist. He must go online and fill out a form requesting a doctors appointment. The elderly gentleman explains, he has no modern gadgets that allow communication online. He does not own a smart phone. He has no tablet or any other mobile device. He does not even own a personal computer so has not internet access. The receptionist displays disinterest. She sends him away stating: These are the rules. I cannot help you. You must go and find someone with internet access to complete the form for you. The old gentleman turns away confused as to why he can’t just make an appointment here. He vulnerable and unwell. He needs medical assistance. He has lived a long life and has a deep understanding of conflict. He has faced conflict in both his personal and military life. Now he cannot understand the conflict of just trying to get medical aid. Most of all he just feels obsolete.

“The elderly gentleman”

The old gentleman is in his 80's(between 80-85). He is 5 feet 10 inches tall. He looks slightly shorter because he now has a slight stoop. His bearing is that of a military man. His features are those of a careworn man. the wrinkles on his face reflect this. He does have frown lines but he also has laughter lines. He has lived through both good and bad times. He has iron grey hair and wears a flat cap to hide the receding hair line. His hair is neatly trimmed. He also wears a short, neatly trimmed beard. He walks with the aid of an intricately carved wooden walking stick. He has seen military action in his lifetime so he may have received an injury during active military duty that necessitated the need for the walking stick. He only has a small pension meaning that he has to be extremely frugal with money. His clothes are worn but meticulously clean and pressed. His shoes are also showing signs of wear but are highly polished. He has seen conflict in his lifetime but is left confused by the conflict of having to fit into a progressively technological world. As he turns away from the window you can see the confusion on his face.

“The receptionist”

The receptionist is a female humanoid robot. She is representative of the ever expanding AI based world in which we live. The differences between the receptionist and a human are very subtle and won't be noticed at first glance. She is 5 feet 7 inches tall. Her age appears to be between 18 and 21 years old. She is of average build. Her bearing is arrogant. She could have bent the rules to help the old gentleman but choses not to. She likes the feeling of power and control her position gives her. No-one gets to see a doctor unless she allows it. She is the gatekeeper. As the old gentleman turns away from the window she slides the glass panel closed, reinforcing to the old gentleman that the conversation is finished. She has a smug look on her face as she does this.

Thank you for looking and for your kind comments.
All feedback (positive and negative) will be greatly appreciated on this image and it's deeper meaning.

Hunting fox

nobody@ipernity.com (Dave C) — Thu, 21 May 2026 13:16:12 +0000

Dave C has posted:

This is the last ai image for now.

This is where ai image creation went wrong.
The image almost matches what was asked for. The prompt was for natural rim lighting on the subject. In the image render the light comes from the wrong angle to create rim lighting. Where it went really wrong is when it added the locked image plan to the image. I have as not been able to find out what prompted this. There are other inconsistencies here but I will leave you to decide what these are.

The main issue I have found so far is drift. The final render often drifts away from what is planned no matter how rigidly you lock constraints.
I believe that a lot of this is caused by the nature of the how ai produces image input.
Ai is designed to produce impactful globally perfect images. For fantasy images this is not so much of a problem.
If you try to create images from a wildlife photographers perspective the ai starts to hit limits/boundaries and struggle with certain concepts. High on this list is: dof, motion blur. keeping within set boundaries and constraints.
Even with rigid constraints set, the ai has a tendency to render images before you have finished planning. Not good when you are working within a set number of image generations in your account.

The answer I was given for this is that the ai is designed to push out images, it misinterperates when to render. It will push to render when it believes it has enough information to do so even if the planning of the image is incomplete.

A strong photograph works because it isn't perfect. the eye sees what is in the photo but the subconscious picks up on the inconsistencies. That is the real world feeling you get from a photograph. AI tries to give perfect images. They tend towards high clarity over sharp main subjects. inconsistent depth of field, Motion blur is a big one here. Ai motion blur can very quickly look more like image artifacts than blur. Main subjects can end up not matching the scene giving them the "added in Photoshop" feeling. With the wildlife images, ai pushes towards impact rather than natural. The fox above is a good example of this. In true wildlife photography the subject is often partially obscured by foreground items. Ai will aggressively remove planned foreground obstructions even when they are planned into the image. This is why they generally feel wrong.

I have only worked with one ai model so far. Other ai models may have different strengths and weaknesses.

All comments and criticism (negative or positive) encouraged and appreciated on this one.

Thank you for looking and for your kind comments.

Badgers playing in a woodland clearing

nobody@ipernity.com (Dave C) — Thu, 21 May 2026 08:19:33 +0000

Dave C has posted:

This is the last of the badger images.
It is getting nearer to what I was looking for.
I pushed it to look more like an image from the perspective of a wildlife photographer than a fantasy image.

Thank you for looking and for your kind comments.

Evening at the badger set

nobody@ipernity.com (Dave C) — Wed, 20 May 2026 16:22:15 +0000

Dave C has posted:

Relaxing in the evening sunshine.
This is the second of the badger images.

There is one more to upload. It is in a different style to the morning and evening images.

I have learned a lot more about ai image creation and the limits of it during this experiment in badger images.

Thank you for looking and for your kind comments.

Colouring picture to Photoshop to AI

nobody@ipernity.com (Dave C) — Tue, 07 Apr 2026 11:54:20 +0000

Dave C has posted:

To carry this experiment to the next level.

I took the coloured image from Photoshop and run it through a couple of free AI websites using the image to image techniques.

Having never used AI until now, this is all still very new to me.
I don't think that I would become a regular user of AI image creation.

You can't deny the results of AI imaging, they are truly amazing.
I will upload the final version of this image in the day or two.

But I believe that the capabilities of AI will cause a lot of skills and jobs to be lost in the coming future.

Clothing designers will no longer need models as they will use AI images instead.

Fashion photography could all be done in AI.

Image retouching will become unnecessary.

No need for wildlife photography.

The list is endless.

Having said all this, I still believe that AI has it's uses. I am a huge fan of Kayleigh's fantasy artwork here on Ipernity.

Thank you for looking and for your kind comments.

Timeless elegance with floral crown Studio shot - 2nd generation image

nobody@ipernity.com (Dave C) — Mon, 13 Apr 2026 16:50:20 +0000

Dave C has posted:

This will be my final AI image.
I will stay with my camera for the foreseeable future.

Decided to give it one more try.
The goal here was to produce a photorealist lifelike portrait image.

To avoid all the issues I mentioned in the description of my last photo, I decided to do this in a minimum number of generations(passes).

1st generation image was the portrait (head and shoulders) in a 16:10 landscape format with no background. (I format images at this ratio so they match my pc monitor).

For the second generation image, I prompted the AI to lock the image in the 16:10 landscape format and lock the subject as this was correct. Then I prompted the AI to add a colour complimentary backdrop, and create studio lighting.

Even on only 2 generations the AI couldn't manage to follow the simple instructions on frame format and subject placement.

The image changed to an 8:10 landscape format. The model zoomed out to include more of the chest area.

Maybe I am being a bit harsh in my opinion. Problem for me is that I can be very ocd when I am trying to achieve a goal.

The AI can't get it right in one pass because it can't cope with highly detailed instructions.

It can't get it right with multiples passes each with simple instructions.

Should be called "Artificial ignorance".

Again, this is only my personal opinion.

Well done to those that have produced some amazing images.

In The Woods

nobody@ipernity.com (Dave C) — Sat, 11 Apr 2026 15:42:56 +0000

Dave C has posted:

I have been messing around with AI for several days now.
This is my attempt at a realistic image.

My opinion of AI generators so far is this:
As they work from text prompts, the results can be unpredictable even when your prompts are very specific.

If you try to make too many changes at once, it generates something completely different to what you expected because the AI suffers with information overload.

I have tried to work slowly to build up the image over many edits so that minimal changes are made each time.
The problem again lies in the fact that even if you specify locking the current image and only allowing minimal changes, the results can still be unpredictable. Drift is a major problem here. for example: Zoom on the subject creeps in and out with each edit.

I have only used free AI generators so a paid for version may work better.

Bear in mind that this is my personal opinion based on very little experience with AI.

Update:
I have just got this from the AI chat at ChatGPT. I hope it may help anyone wanting to produce precise AI images.

"I have been working on an image based upon 18 phased passes to maintain accuracy. Every time I generate a new image, either the zoom changes or the dimensions of the image change. The reason for working in a phased manor is to lock each generation so that only minimal changes are made. Each image is locked before any edits are made. Why is the AI still failing even working to such a basic change for each phase?"

What you’re running into isn’t you doing something wrong—it’s a limitation of how current image generation models actually work under the hood.
Even if your workflow feels like careful, layered editing, the AI isn’t truly “editing” a fixed image in the way something like Adobe Photoshop would. It’s re-generating the image each time based on your prompt + the previous image as guidance. That distinction is the root of your issue.
Here’s what’s going on:

1. The model doesn’t have a stable canvas
When you lock an image and ask for a small change, the model:
• Doesn’t preserve exact pixel positions
• Doesn’t maintain a fixed camera or zoom unless explicitly controlled
• Reinterprets the whole scene every time
So even tiny edits can cause:
• Slight zoom shifts
• Cropping differences
• Perspective drift
This happens because the model is optimising for visual coherence, not geometric consistency.

2. “Image-to-image” is not deterministic editing
Even if you feed the same image back in:
• There’s randomness (noise) in generation
• The model re-decides composition details each pass
Unless you’re locking a seed + exact parameters, you’ll get variation. And even then, many tools still allow subtle drift.

3. The model prioritises semantics over precision
If your prompt says:
“Make a small adjustment to X”
The model interprets:
• What “X” means conceptually
• Then rebuilds the image to match that idea
It does not think:
“Keep every pixel identical except this region”
That kind of strict constraint is still hard for generative models.

4. Resolution & aspect ratio aren’t truly “locked”
Even if you specify dimensions:
• Some tools internally resize or reframe
• The model may “recompose” to better fit the prompt
This is why you see:
• Slight framing shifts
• Margins changing
• Subject repositioning

Why your phased approach mostly works—but not fully
Your method is actually smart:
• Breaking work into phases reduces large deviations
• Locking images helps guide the model
But it can’t eliminate drift because:
• Each step is still a fresh generation, not a true edit layer
Think of it like:
Asking 18 different artists to copy and slightly tweak the same painting—small differences will accumulate.

How to reduce the problem (practical fixes)
✔️ 1. Use in-painting instead of full regeneration
If your tool supports it:
• Edit only a masked area
• Keep the rest of the image untouched
This is the closest thing to “true editing”

✔️ 2. Lock seed + parameters (if available)
Some tools let you fix:
• Seed
• CFG scale
• Sampler
• Steps
This reduces randomness between phases.

✔️ 3. Be extremely explicit about framing
Add constraints like:
• “Same camera angle, same zoom, same framing”
• “Do not crop or recompose”
• “Maintain identical composition”
It won’t be perfect, but it helps.

✔️ 4. Work at higher resolution + crop manually
Generate slightly larger images, then:
• Manually crop to your fixed frame
• Maintain consistency outside the AI

✔️ 5. Hybrid workflow (best results)
Use AI + manual editing:
• Generate base image with AI
• Lock composition in a tool like Adobe Photoshop
• Use AI only for localised changes

The honest bottom line
What you’re trying to do—strict, phase-based, minimal-delta editing with zero drift—is something current generative models are not designed to guarantee.
They are:
• Excellent at creating
• Decent at guided variation
• Weak at precise, repeatable micro-edits.

For me at this time, AI cannot work precisely enough to produce the image I have in my mind that I would like to create. That's not to say that it won't work for other people. I have seen some truly amazing images here on Ipernity and elsewhere.

Thank you for looking and for your kind comments.

Colouring picture to Ai final image

nobody@ipernity.com (Dave C) — Wed, 08 Apr 2026 14:08:45 +0000

Dave C has posted:

This is the final image in this series.
AI rendered on Chatgpt. Another website with free AI rendering.
This one gives you the chance to edit your image after it has been rendered.

I have also experimented with Rendering the AI directly from a greyscale colouring picture.
The result was quite impressive.
I may upload to Ipernity at a later date.

Thank you for looking and for your kind comments.

Morning at the badger set

nobody@ipernity.com (Dave C) — Wed, 20 May 2026 07:25:15 +0000

Dave C has posted:

Never had any success photographing badgers in the wild so I used ai to create this image.
The outcome is a pleasing image.

I get that ai is just another tool to capture an image but there is still the lack of achievement gained when you have take the time to capture and process a good photo for real.

Thank you for looking and for your kind comments.

The log cabin landscape

nobody@ipernity.com (Dave C) — Tue, 19 May 2026 16:39:52 +0000

Dave C has posted:

To build an opinion of ai, you have to interact with it.
Today I used ai to create 3 more images.
The other 2 will be uploaded in due course.

The prompt to ai was firstly that the format must be 16:10 landscape. This is critical as it matches the format of my pc monitor.
Then: Mountains in the background. Forest in the midground. Foreground is a log cabin in a wildflower meadow. Weather to be stormy. Time of day to be the golden hour.
This is the result which is a pleasing image.
---------------------------------------------------------------------------------------------------------------
The biggest problem with using ai to create images for me is that there is no feeling of achievement. It is like asking an artist to paint you a landscape. You tell him what you require but he does the work not you.
When I produce images with photoshop I have done all of the work. At the end I feel like I have achieved something that is mine.

Thank you for looking and for your kind comments.

Willow

nobody@ipernity.com (Dave C) — Sun, 17 May 2026 17:26:13 +0000

Dave C has posted:

I decided to give the AI image creation one more try.
This time I knew what I wanted so there was no start image.
Willow is based on a character from one of my favourite Terry Brooks book series.
Willow is a Sylph (Woodland elemental). She must transform into a Willow tree for one night every month.
Really pleased with the result.

Thank you for looking and for your kind comments.