Saturday, October 7, 2023

Tensor Art AI

 

I recently came across yet another AI imaging app - there seems to be a new one every week - named Tensor Art.  This one, at least on first impression, seems better than most I've tried and somewhat resembles Leonardo AI, though Tensor appears to offer even more features for fine tuning images including inpainting, the ability to add Loras (a premium feature on Leonardo) and specify seeds, a choice of adetailer models and sampling methods, a hires fix that not only offers a choice of sizes but also of the number of steps to achieve that size.  And Tensor, of course, also offers the more commonly found features such as a choice of aspect ratios and image-to-image prompts.  For a free app (a set number of credits are assigned daily with each image generation costing a single credit) it's really very impressive.  There isn't even the need to set up an account - one simply logs in using one's email address,

Tensor's image quality, if not quite up to Midjourney standards, is still very good,  Below are several images generated using one of the app's platform models.

For those interested in taking the app for a test run there's a useful YouTube tutorial by Making the Photo that concisely explains all Tensor Art's features in less than 15 minutes.



Thursday, October 5, 2023

Leonardo AI Now Offers Loras


Leonardo AI, an excellent AI imaging app. generously offers a free tier of service to its users in the form of 150 credits a day with which to generate images.  The app is, however, a commercial venture and in order to increase its revenue stream has hit on the idea of adding new "premium" features, seemingly one every few weeks, that are only available to paying subscribers.  I can't really fault Leonardo for doing this since its subscription rates are reasonable enough (though they certainly have to be to remain competitive with Midjourney's) and the premium features themselves, such as Alchemy and PhotoReal, are generally very useful.

Leonardo's latest new paid feature is Elements, its own brand of LoRa (Low Rank Adaption) models that have the ability to greatly enhance the appearance of generated images, either on their own or in conjunction with the app's platform models.  Leonardo describes the new feature as follows:

"We are thrilled to announce that Leonardo Elements; our proprietary custom implementation of LoRA (Low-Rank Adaption) models are now available to use on the platform.

Taking advantage of Elements is easy; in order to get started hit the new ‘Add Elements‘ button under the prompt generation input:

Select from Elements you would like to use for your generation. You can select up to four Elements to combine, or use one on it’s own. Find and experiment with different styles, find your favorites - and go wild!

Once your selection is complete, you can adjust the weights for each Element to fine-tune your results. Adjust weights to get the perfect balance for your generation."

In addition to this introduction, there is a useful if overly long YouTube tutorial by Wade McMaster that tells everything there is to know about Elements and then some.  The video would actually have been better at half its almost 10 minute length.  As it is, there are a few too many samples shown and the accompanying commentary would have benefited by being more concise.

Leonardo has come a long way in only the past few months and now is fairly comparable to Midjourney in the quality of the images its generates and the features it provides.  It's the one app I feel makes it worth upgrading to a paid subscription, both for the aforementioned quality of its images and for its ease of use,  and I will probably do so sometime in the near future when I have more time.

Tuesday, October 3, 2023

Nikon Announces 135mm Plena Z-Mount Lens

Nikon has just announced the release of its 135mm, f1.8 full frame S Plena Z-mount lens which it has been teasing for the past several weeks.  In case you're wondering, the "S" in Nikon's S line apparently stands for "Supreme Optics and Design," and judging from its specs and the sample photos on Nikon's website this quality optic certainly appears to merit that designation.

135mm has always been every camera manufacturer's most commonly used focal length for portrait (i.e., moderate length telephoto) lenses, but unless the sample photos lie there's nothing common about this optic that boasts the following features:

  • Aspherical design;
  • Extra-low dispersion (ED) glass;
  • Short-wavelength refractive (SR) glass for precise chromatic aberration compensation;
  • Meso amorphous coating;
  • Arneo coating;
  • Stepping motor;
  • Auto-priority manual mode;
  • Internal focusing that prevents the lens changing in size during auto-focusing;
  • Electromagnetic diaphragm mechanism

Impressive as these specs are, it's the quality of the sample photos that is truly outstanding.  The bokeh produced by the Plena is nothing short of amazing, and the color rendition is nearly perfect.

At a list price of $2,496.95 the Plena is certainly not for every photographer, but if it lives up to its advertising I'm certain it will become a must-have for every serious portrait and wedding professional as well as those bokeh enthusiasts willing to pay a premium to achieve effects such as those shown in the sample photos.

Monday, October 2, 2023

Adobe Firefly

As I'm always interested in experimenting with new AI imaging apps, I recently turned to Firefly after having received an email from Adobe last week inviting me to try out the newest version.  I had actually taken a look at the app when it had first been announced by Adobe several months ago and had not been particularly impressed by what I had seen.  (See my March 23 post.)  The text-to-image results I obtained today, though, were far superior to what I had seen earlier in that they were much more photo realistic.

In addition to text-to-imaging capabilities, Firefly has several intriguing new features - generative fill, text effects, generative recolor, sketch-to-image, and 3D-to-image - and has several more in the works, including personalized results, text-to-vector, text-to pattern, text-to-brush, extend image, and text-to-template.  I haven't yet tried any of these myself but they certainly appear promising.

As far as the text-to-imaging results I obtained on my latest try, and with which I was fairly well pleased, I am posting a few samples below.  (Note that all the human figures generated by Firefly were minority members even though I had not requested this.)  As a prompt, I used text generated by ChatGPT originally intended for use in another AI imaging app.  The prompt reads as follows:
"In the dystopian metropolis of NeoCity, a weathered and weary replicant, known only as X42, finds solace in recounting its extraordinary experiences to a disenchanted group of rebels. Its voice resonates with both longing and resignation as it speaks: 'I've seen things you people wouldn't believe. Attack ships on fire off the shoulder of Orion. I watched C-beams glitter in the dark near the Tannhäuser Gate. All those moments will be lost in time, like tears in rain.'"
Whatever its literary merits, the prompt successfully gave me the cyberpunk vibe which I was seeking.

As I was downloading the images generated by Firefly, I noticed that for each a label appeared that read: "content credentials applied."  I then clicked the Learn More button to discover exactly what Adobe meant by "content credentials."



Saturday, September 30, 2023

DxO Releases FilmPack 7

 
In addition to the recent release of PhotoLab 7 DxO has also announced that of FilmPack 7, the app whose purpose has historically been to recreate the appearance of various brands of film, both those that have been discontinued and those still on the market.  I'm a current user of FilmPack 6 and will almost certainly eventually upgrade to the new version if the price is right.  I deliberately put in this hedge regarding price because there are unfortunately not that many new features in version 7 compared to those that came with version 6.  Those that are included can be summarized fairly quickly:
  • Luminosity masks to provide greater control when editing within the app;
  • Time Machine has been updated with 22 new images generating 26 new presets;
  • Five new film simulations and 20 new digital renderings.
I believe this is the first time digital renderings have been included in the simulations, an indication that the app is moving away from its original purpose of recreating analog looks.  This was doubtlessly necessary since by now FilmPack has a fairly full roster of nearly every film ever manufactured.

FilmPack 7 is available for download for $139, or $79 for the upgrade.

Friday, September 29, 2023

DxO Releases PhotoLab 7

DxO has just announced the release of PhotoLab 7, the latest version of its flagship RAW processing software.  Until last year when On 1 released PhotoRAW 2023, the successive versions of PhotoLab had always been my go-to app for RAW processing.  That's no longer the case.  PhotoRAW's 2023 version was so impressive that it really became for me a one-stop app for RAW processing, photo editing, and cataloging; and the soon to be released 2024 version promises to be even better.  For that reason I doubt that I will upgrade from PhotoLab 6, though I most definitely will keep it on my computer.  Then too, the new features listed by DxO for PhotoLab 7 are more incremental than revolutionary.  For the record, these include:

  • Color calibration;
  • Local corrections palette;
  • ColorWheel correction availability for local adjustments;
  • DxO styles and renderings;
  • B&W workflow;
  • Channel mixer (also available in in FilmPack 7)

That's not to say PhotoLab is not an excellent app.  It certainly is, and there may be many photographers who feel more comfortable using it than PhotoRAW.  In any event, I think both apps are far more advanced than Adobe Lightroom.

PhotoLab 7 is available for download for $229, or $109 for the upgrade.

Thursday, September 28, 2023

Panasonic Announces the Lumix G9II

As an owner of the Panasonic Lumix G9 I was intrigued by the announcement of its successor, the G9II, though I honestly don't feel any great need to upgrade at this time.

Rather than regurgitate the camera's specs, which are readily available elsewhere, I'd refer the reader to DPreview's initial take on what the G9II has to offer.  Perhaps the most important improvement from the older model is the inclusion of on-sensor phase detection, previously available only in Panasonic's full frame S line, that should make correct focusing far easier.  And then too the camera, though like its predecessor marketed primarily to stills photographers such as myself, apparently evidences far improved video capabilities that in some respects surpass those of the GH6.  Other innovations, however much they may be hyped, appear more incremental than revolutionary.

The G9II is available for preorder from B&H at a price of $1,897.99 (body only) is expected to begin shipping in early November.

Wednesday, September 27, 2023

On 1 Announces PhotoRAW 2024


On 1 has announced the latest version of its flagship app, PhotoRAW 2024.  With an expected release sometime in November, the latest version doesn't seem as much a gamechanger as last year's release, but it still seemed to me to be well worth preordering.  

The big feature on PR2024 - and it is major - is what On 1 calls AI Brilliance, which promises to in one shot automatically make all required adjustments to the RAW image being processed as well as automatically applying local adjustments.  It sounds too good to be true, but I'll withhold judgment until I've has a chance to try it for myself.  The other big change for  Adobe users - which I most definitely am not - is that On 1's plugin series is now incorporated in the Max edition of PR2024 allowing the app itself to be used as a plugin in both Lightroom and Photoshop.  (Though not specifically mentioned, I would assume this is also true of Photoshop Elements.)  The remaining changes to PR2024, while useful, are definitely not as revolutionary.  For the most part, they're limited to increased speed, an updated interface that allows users to more easily switch from browse to edit modes, a revamped layers pane that now allows text to be saved in layers, improvements in speed and cataloging capabilities (including keyword enhancements), and the inclusion of 100 new "curated" presets to effects.

After leaving Adobe, I search for quite a while for a satisfactory RAW processor and photo editor.  I had pretty well settled on DxO PhotoLab 6 and Skylum Luminar Neo.  When On 1 released the 2023 version of PhotoRAW, however, I was impressed enough to make the switch and now use PhotoRAW almost exclusively for processing and editing.  I'm confident I will be able to continue doing so with the 2024 version.

Tuesday, September 26, 2023

Leonardo AI PhotoReal

 

I posted on Saturday about a trial offer from Leonardo AI that would enable me to test drive the new PhotoReal platform model that is normally available only with a paid subscription.  Above are two images I obtained during my trial.  I was very pleased with the results I obtained.  If the images are not yet quite at the level of Midjourney they're certainly within shouting distance.  Shown below is the text prompt I used to generate these images.

"low key elegant depiction photographed on analog color film in muted colors of beautiful model dressed in futuristic designer clothing and stiletto heels posing for high end fashion advertisement on balcony of expensive high rise apartment in rain drenched neon lit Tokyo"

For comparison's sake, I've shown at bottom two images I obtained using the same prompt with a non-premium platform model - Absolute Reality v1.6.  They're good, at least in my opinion, but obviously not as good.  Whether the difference is worth paying for is an individual user's personal choice.  Note that the premium Alchemy module was turned on in both sets of photos.


Monday, September 25, 2023

Topaz Labs Release Photo AI 2.0

 

A year or two ago Topaz Labs instituted a new policy regarding upgrades.  Rather than entitling customers to free lifetime upgrades on the apps they had purchased, those customers were now informed that they were only entitled to free upgrades for the twelve-month period immediately following purchase, after which they would have to pay to have their upgrade license renewed for another year.  I didn't think this was a particularly good way of doing business, but I so far have not suffered from it.  Eleven months after I had purchased Video AI 2.0 Topaz released version 3.0 that contained a number of new and improved features, and I was able to download it at no charge.  More recently, only a few days before my license was due to expire on Photo AI, Topaz released version 2.0 and I was again able to download it at no charge.

It's lucky the upgrade was free because I certainly never would have paid for it.  I don't think I was ever so disappointed in an app as I was in Photo AI.  Unless I'm doing something very wrong - and I strongly doubt that - the app is too unwieldly to be truly useful and has never become part of my workflow.  The biggest problem I've encountered is that the vaunted Autopilot simply doesn't work properly.  It seems whatever image I put up, the Autopilot invariably detects that denoising is required even though almost all my images are shot on up-to-date cameras at relatively low ISO's.  The Autopilot never flags sharpness or any other needed corrections on its own with the result that I am required to go in and make corrections manually.

When I have time I'll try exploring version 2.0 in greater depth, but I honestly don't have any great hopes for it.  In the meantime, I still regularly use the same company's Studio 2, discontinued  in 2020, particularly the AI Clear and Basic Correction filters.  I would much have preferred if Topaz had released a Studio 3 version, but that doesn't look likely to happen.

Saturday, September 23, 2023

Leonardo AI Free Trial Offer

I received an email earlier this week from Leonardo AI offering an opportunity to participate in a free five-day trial of its Alchemy tool (currently in beta) and its new AI model PhotoReal.  It's an offer I'm definitely going to take advantage of - I've only held off up to now because I wanted to be able to put aside sufficient time to experiment - and will probably do so in the next few days.  While Midjourney may still be the leader when it comes to AI imaging apps, others are fast closing the gap.  I had a brief taste of Alchemy as a freebie when I first set up an account on Leonardo and was impressed by how greatly it improved image quality even when using the same text prompt.  I'm hopeful that PhotoReal will bring me even closer to Midjourney-level results.  If it does I'm almost certain to set up a paid account next month after I've gotten some important personal business out of the way.

The same email that announced the free trial also announced a contest on social media in which participants "Blend two unlikely subjects in spectacular fashion!" by using Alchemy or PhotoReal to "create a masterpiece of fusion."  The grand prize will be a huge 30,000 tokens.  To put that in perspective, I only receive 150 tokens per day on the free level to which I currently belong.

I don't know if I'm going to bother taking part in the contest, but hopefully I will be able to post here in coming days images I've obtained using Alchemy and PhotoReal.  Stay tuned.

Thursday, September 21, 2023

Nikon Releases the Zf

 
Nikon may have moved on years ago from producing SLR film cameras, but it certainly has never lost its love of their design.  And why shouldn't it indulge in a bit of nostalgia?  After all, it was the fantastic success enjoyed by the Nikon F when first introduced in the late 1950's that made the company into the behemoth it is today.  And so it should come as no surprise that Nikon has routinely turned out DSLR's (the full frame Df that is now all but discontinued) and mirrorless models (the APS-C Zfc and now the full frame Zf) whose appearance is decidedly retro.  As a long time photographer whose go-to camera for the past 35+ years has always been the F3T (the T stands for my model's titanium body) I've always been intrigued by these cameras and currently own both a Df and Zfc.  It was only natural then that I should be intrigued by the release of the Zf whose design mimics that of the long discontinued FM.

I don't see much point in regurgitating the Zf's specs when they can easily be viewed both on Nikon's own website as well as B&H's, so I'll content myself with a few random observations.

First, the Zf's sensor size is at only 24.5 MP far smaller than that of the Z8 and Z9 and much more in line with the Zfc's 20.9 MP.  For me, this indicates right off the bat that the Zf is not intended for professionals but rather for the so-called "enthusiast" market.  Nevertheless, 24.5 is still a substantial size and should be adequate for most users especially as it is now quite easy to upsize images with third-party apps such as Topaz Labs' Gigapixel and On1's AI Resize.

In many ways the Zf's specs are similar to those found in the Z6 II, but there are some key differences.  For one, its AF detection capabilities go as low as -10EV, a boon for those who depend on autofocus in low light.  The Zf also boasts a fully articulating touch screen which, for me at least, would prove extremely useful.  Also useful for those who like myself suffer from poor eyesight is the subject recognition with 3D tracking and the eye-detection autofocus for both photos and videos.  I also appreciate the inclusion of dual card slots even if one of these supports only micro-size cards.  Of much lest interest to me, however, is the camera's ability to quickly toggle between color and b&w modes.  Though for most of my career I specialized in shooting b&w film, when it comes to digital I always shoot in color and leave the b&w to post-processing apps, of which I've found NIK Silver Efx to be the best.  Finally, I should mention that the Zf's shutter release, like that of my Df, is threaded for the use of a cable release which is what I always use when shooting with a tripod.

The Zf is now available for preorder from both Nikon and B&H for $1,999.95 (body only), though no date I can see has been given when the camera is actually expected to ship.  That seems a reasonable enough price for a full frame camera with the Zf's specs, especially when compared to the double-that-amount price for the recently released Z8 which, however, admittedly contains a slew of professional features not available on the Zf starting with its 45.7 MP sensor.

I probably will purchase the Zf at some future date but see no need to rush out and be the first to buy.  As I'm pretty much retired now I can get along quite well with my current array of cameras and would prefer to wait to see if there turn out to be any quirks that need to be addressed with updates.  There almost always are following the release of any new camera model.

EDIT: According to an update on B&H's website, the Zf will begin shipping on October 31.

Tuesday, August 22, 2023

New Add-Ons for Luminar Neo

 

Skylum has just announced six new add-ons, or editing tools, for Luminar Neo; and I have to admit that I find the selection rather underwhelming, especially as in at least a few cases similar tools are offered in other apps, sometimes for free.  

The new features are as follows:

  • Gen Erase uses AI to remove unwanted elements from photos with a single click;
  • SceneExpand enlarges the dimensions of a photo by adding content to is borders;
  • SceneSwap replaces skies and foreground elements (including, apparently, articles of clothing) with AI generated content;
  • Studio Light adds lighting effects to photos, though in the examples shown these appear to consist of nothing more than unrealistic bands of shadows;
  • Neon & Glow, as the name implies, adds neon borders and glow effects to photos, though the results are distressingly hokey in the examples shown; 
  • Water Enhancer seems to do nothing more than add varying shades of blue color to whatever body of water is shown in a given photo.

For some reason the one-time sale purchase price of the above tools distinguishes between Generative AI Features (GenErase, Scene Expand, and SceneSwap), which are free of access until August 16, 2024, and Platform Features Upgrades (Studio Light, Neon & Glow, and Water Enhancer), which are free of access forever.  The $39 cost is not a particularly high price to pay, but I doubt I'll be making the purchase myself since none of these tools are of any great interest to me and would be of limited use in my current workflow.

Monday, August 21, 2023

Federal Court Rules Against Copyrighting AI Images

 

In a summary judgment ruling, which basically means the case - Thaler v. Perlmutter - never went to trial, the US District Court for Washington, DC has sided with the US Copyright Office in finding that works created by artificial intelligence are not entitled to copyright protection since they have no human author.  Though I am not an attorney and therefore am not qualified to render a legal opinion, it does seem to me that the decision is not as sweeping as it might first appear.  In the Discussion section of its opinion the Court states:

"Undoubtedly, we are approaching new frontiers in copyright as artists put AI in their toolbox to be used in the generation of new visual and other artistic works. The increased attenuation of human creativity from the actual generation of the final work will prompt challenging questions regarding how much human input is necessary to qualify the user of an AI system as an 'author' of a generated work, the scope of the protection obtained over the resultant image, how to assess the originality of AI-generated works where the systems may have been trained on unknown pre-existing works, how copyright might best be used to incentivize creative works involving AI, and more."

Here the Court seems to be leaving the door open to further discussion and different rulings.  The Plaintiff's argument, at least as I read it, was that the AI imaging app he invented should be considered the "author" of the work in question because it autonomously created said work with no human input.  Whether or not the app did manage this, it is certainly not the manner in which most AI images are generated.  Midjourney and Stable Diffusion most certainly do not autonomously spit out images without direction but rather respond to prompts input in text or image form by human users.  Beyond that, the resulting images generated by those apps are very often edited in other third-party apps such as Photoshop, thus providing further evidence of human involvement in the authoring process.

In other words, I believe the Court's decision in this case was based on very narrow grounds and that subsequent cases in which human involvement is clearly demonstrated may eventually cause the Copyright Office to grant protection in such circumstances to the human creators who directed the generation and editing of the AI images in question.

Tuesday, August 1, 2023

Is Stable Diffusion XL Actually a Step Back?

Perhaps it's only been my experience, but after having tried the newly released Clipdrop Stable Diffusion XL 1.0 extensively for the past several days I've found myself sorely disappointed.

One of my favorite methods of image prompting has been to simply to enter a few lines of poetry and then sit back and see what a given AI model makes of them.  As an example, I've in the past had good luck using the opening from John Keats's famous Romantic poem Ode to a Nightingale that reads as follows:

My heart aches, and a drowsy numbness pains

         My sense, as though of hemlock I had drunk,

Or emptied some dull opiate to the drains

         One minute past, and Lethe-wards had sunk:

'Tis not through envy of thy happy lot,

         But being too happy in thine happiness,—

                That thou, light-winged Dryad of the trees

                        In some melodious plot

         Of beechen green, and shadows numberless,

                Singest of summer in full-throated ease.

Whether or not the poetry itself is to one's taste, it's definitely given me some truly inventive images, as can be seen in the example at the top of this post that was generated by SDXL 0.9.  And what's notable is that that older version of the app managed to maintain the same dark style through hundreds of image generations.

Unfortunately, I haven't had nearly the same luck when working with SDXL 1.0.  One need only look at the image below to see the problem.  In place of the edgy images generated by 0.9, I've instead gotten bland generic illustrations better suited to a children's storybook, even though I used the exact same prompt and style (digital art) for both the images shown here.  It's very frustrating, and I've had the same experience with a wide variety of prompts, and not only with poetry either.  It seems the 1.0 AI model is simply not as imaginative as the 0.9, strange as that must sound.

I note that one of the options available when using Clipdrop SDXL 1.0 is the choice of version.  One can opt to go back and once again use version 0.9, but only if one subscribes to the paid Pro plan.  Since the cost isn't all that great ($84 per year) I may eventually decide to go that route.  To me, it would definitely be worth the money if I could once again generate the images I find most appealing.  There are certain other benefits to the Pro plan as well, such as the elimination of watermarks and faster image generation, that make it worth considering.

Friday, July 28, 2023

Is DALL-E 3 About to Be Released?


Back in the fall of last year the buzz in AI imaging, then only beginning to take hold of the popular imagination, was all about Open AI's DALL-E 2.  How long ago that now seems.  Since then, Midjourney has deservedly been been recognized by most as the premiere text-to-image AI generative app while DALL-E 2 has been largely forgotten.  It's no surprise.  Even at the height of its popularity its imagery seemed, to me at least, markedly inferior to that which I obtained from it's main rival, Stability AI's Stable Diffusion.

In spite of the above, it seems Open AI has not yet given up on DALL-E and, at least according to a YouTube video I recently viewed, is preparing to launch version 3.  The original video seems to have been removed since then, or I'm simply unable to locate it, but I remember thinking when I viewed it that the "leaked" images shown were not much better than those I'd seen generated by version 2.  At any rate, there is no official word as of this writing from Open AI on any plans to update DALL-E.  I'll of course post here if I hear anything definitive.

Tuesday, July 25, 2023

ChatGPT Formula for Generating Prompts

 

I recently came across a YouTube video by a creator named Monice in which she gave a formula that would enable one to use ChatGPT as an image generator for Midjourney.  The given formula to be entered into ChatGPT reads as follows:

"Here’s a Midjourney formula for you to understand.

‘/imagine prompt: [Image we are prompting], [5 descriptive keywords], [Camera type], [Camera lens type], [time of the day], [focal distance], [style of photograph], [by photographer name] –ar x:y Reply ‘yes’ if you understood the formula."

I decided to give it a try and ChatGPT quickly furnished me with the following parameters:

"/imagine prompt: [Image we are prompting], [5 descriptive keywords], [Camera type], [Camera lens type], [time of the day], [focal distance], [style of photograph], [by photographer name] –ar x:y

[Image we are prompting]: This is the main subject or content of the photograph that will be used as a basis for the prompt.

[5 descriptive keywords]: Five words that describe the main elements or characteristics of the image.

[Camera type]: The type of camera used to capture the photograph.

[Camera lens type]: The specific lens used on the camera to capture the image.

[time of the day]: The specific time of day when the photograph was taken (e.g., morning, afternoon, evening).

[focal distance]: The distance at which the camera's lens is focused when capturing the image.

[style of photograph]: The artistic style or approach used in the photograph (e.g., abstract, portrait, landscape).

[by photographer name]: The name of the photographer who took the picture.

–ar x:y: The aspect ratio (ar) of the photograph, represented by 'x:y'"

I in turn provided the following information:

"Here are the specifics for each parameter as numbered above: 1. image we are prompting = A depressed young woman sits on wet floor in an unfurnished darkened basement and stares forlornly at a tiny skylight high above her head; 2. five descriptive keywords = depressed, melancholy, darkened, forlorn, muted; 3. camera type = nikon df dslr; 4. lens type = nikon 23mm wide angle lens; time of day = 4pm EST; focal distance = 5 feet; style of photograph - artistic vintage color; name of photographer = paola m franqui; aspect ratio = 1620 px wide by 1080 px high"

ChatGPT then issued the following prompt:

"A depressed young woman sits on the wet floor in an unfurnished darkened basement and stares forlornly at a tiny skylight high above her head. Keywords: depressed, melancholy, darkened, forlorn, muted. Camera type: Nikon DF DSLR. Lens type: Nikon 23mm wide-angle lens. Time of day: 4 PM EST. Focal distance: 5 feet. Style of photograph: Artistic vintage color. By photographer Paola M Franqui. –ar 1620:1080"

Since I do not currently have an account on Midjourney, I instead entered the prompt into Clipdrop Stable Diffusion XL which generated four images from it, two of which are shown here.

I thought the images were very much in line with what I had been looking for, though I don't know if it really had been necessary to go through ChatGPT to get them.

Monday, July 24, 2023

Using Poetry for AI Prompts

 

As I'm always in search of new ideas for AI imaging prompts, it recently occurred to me to simply enter random lines of poetry and then wait to see what a particular AI model would make of them.  Rather than work from masterpieces of literature, I instead focused on poetry that contained the most vivid imagery.  Accordingly, the two images shown here were generated by Clipdrop Stable Diffusion XL from the opening lines of the 1912 poem Medusa by Clark Ashton Smith, always one of my favorite Weird Tales authors of whom no one less than H.P. Lovecraft once wrote: "In sheer demonic strangeness and fertility of conception, Clark Ashton Smith is perhaps unexcelled."  And on that score who would know better than Lovecraft?

Medusa itself is contained in The Star-Treader and other Poems, available as a free download from Project Gutenberg, and the lines in question read as follows:

"As drear and barren as the glooms of Death,

It lies, a windless land of livid dawns,

Nude to a desolate firmament, with hills

That seem the fleshless earth's outjutting ribs,

And plains whose face is crossed and rivelled deep

With gullies twisting like a serpent's track.

The leprous touch of Death is on its stones,

Where for his token visible, the Head

Is throned upon a heap of monstrous rocks,

Grotesque in everlasting ugliness,

Within a hill-ravine, that splits athwart

Like some old, hideous and unhealing scar."

Pretty strong stuff, isn't it?  And perfect for the use to which I put it.

My college degree was in English lit; and so I have, even many years later, a wide familiarity with world poetry from which to draw future inspiration.  It will be interesting, to me at least, to see what I can come up with.

Friday, July 14, 2023

Is Lexica the Best AI Imaging App?


While seemingly every time I sign on to YouTube I'm bombarded with buzz on the latest "hot" AI imaging apps - Midjourney, Leonardo AI, and Stable Diffusion XL, for example - one that I almost never hear mentioned is Lexica;  and that's rather strange considering the superior image quality this online app invariably provides.

Though prompting takes a bit more effort than is required with Leonardo AI, Lexica is still extremely easy to use, especially as users are not tied to the Discord app as they are with Midjourney but can generate images directly from Lexica's website in an amazingly short time.  And the pricing, similar to Midjourney's, is definitely reasonable.  Lexica may be stricter about forbidding questionable material, but since I never seek to create any NSFW or even suggestive images that's not really a problem for me.  Most importantly, the images generated by Lexica are, if anything, of even higher quality than those that can be acquired from Midjourney.  Just look at the two examples I've posted here and judge for yourself.

The bottom line, for me at least, is that the AI imaging app that's most worth a paid subscription, at this point in time at any rate, is Lexica.

Monday, July 10, 2023

How Good Is Leonardo AI?


I recently discovered a new text-to-image AI generator named Leonardo that's free to use, though there are paid pricing tiers available.  I don't think most users will need to go the paid route, however, since everyone is given 150 credits each day.  Even though several credits are used for each image generation and more for features such as upscaling, I still found I could get a good number of images without exhausting my daily quota of credits.  But those who feel the need for a larger number of credits, the ability to keep generated images private, and more control using such features as Alchemy, now in beta, which offers a number of style variations and generally results in higher quality images, may well want to consider upgrading to a paid plan.

The great advantage to Leonardo for those who do not have extensive experience with AI generative models is its ease of use, especially when compared to such apps as Midjourney which require a certain level of expertise.  At the most basic level all one need do is choose a model - there are several featured platform models as well as an extensive library of community-created models - that will determine the general look of the generated image(s), choose between Leonardo Style (which will usually result in greater realism) and No Style, and enter a prompt.  Leonardo will do the rest.  It even has a prompt generator that will improve on a submitted prompt within the image generation window itself.

As mentioned, there are a number of features available to those who want more control over their images.  In regard to the image generation itself, one can first of all (at a cost of extra credits) choose the output size and the number of variations.  For myself, I've found two variations on each image and the default 768x768 size to be sufficient.  If I need a larger size I can always upscale later in Topaz Labs Gigapixel.  

Even without the afore-mentioned Alchemy, one can the choose between Leonardo Style and no style at all.  Prompt Magic is turned on by default and its guidance set to .4.  I haven't had a chance to fully explore this feature myself but it is defined quite simply as follows: "Prompt Magic enables higher prompt adherence and greater deviance from training data."  High Contrast is also turned on by default.  A nice touch is the ability to enter negative prompts and thereby exclude from generated images items one does not wish to see in them.

Beyond this, there are several features one would not expect to find in a free app.  A Guidance Scale, for example, determines how closely a given prompt should be adhered to when generating images.  One can also upload images for use in image-to-image prompting while ControlNet "allows you to influence your generations with input images."

One can also edit within the generated image canvas itself, allowing one ""to visually correct and refine details, modify styles, seamlessly extend content, create composite artwork and striking photomontages through inpainting and outpainting."

Finally, and perhaps most importantly, Leonardo offers users the ability to train their own models in only a few clicks.

As far as image quality, Leonardo provides excellent results, though I believe that, unless one has access to Alchemy, Midjourney, Lexica, and the new Stable Diffusion XL may hold a slight advantage.  I did try using Crisp HD Upscale on a few photos, but it didn't really seem to appreciably improve the overall image quality.

In summary, I think that once Alchemy comes out of beta, Leonardo will be a viable alternative to Midjourney as far as quality is concerned.  It's certainly easier to use - one is no longer tied to the Discord server for one thing - and the pricing on the paid tiers is comparable to what Midjourney charges.

Thursday, July 6, 2023

Midjourney 6.0 Arriving Soon?

According to an article in Forbes, which in general is a fairly reliable source of information, AI imaging app Midjourney will release version 6.0 before the end of July.  The article unfortunately did not contain any information on what features the new version may offer.  It did, however, describe in a fair amount of detail two new features in the current version 5.2 and also posted examples to illustrate the results obtained with them.

The first of these new features is the ability to create panoramic images.  According to Forbes, it works as follows:

"When a regular image is upscaled to full resolution, the user gets the option to pan the image left, right, up or down. This effectively extends the image in the chosen direction, with the AI filling in the empty space.

Images can continue to be extended beyond one iteration, although the wider and wider aspect ratio becomes problematic.

Users can also enter Midjourney’s Remix mode and change the prompt at the same time as extending the image, effectively letting them create panoramic photo stories."

The second new feature is Weirdness.  As the name would imply, this feature allows users to dial in various degrees of "weirdness" into generated images on a scale of 0 to 3,000.  How useful Midjourney subscribers will find this feature will obviously depend on their individual styles of creativity.  Those who favor a straightforward approach to image generation will most likely opt not to work with it that often while those seeking to create the unusual and fanciful may find it quite helpful in achieving their goals.

Wednesday, July 5, 2023

Clipdrop Uncrop

 

In a previous post I mentioned that Stability AI had acquired a suite of online apps, the Clipdrop series, as a result of its acquisition in March of Init ML.  Recently I decided to experiment with one of those apps, Clipdrop Uncrop, which as far as I can make out is very similar in purpose to Adobe Photoshop's Generative Fill and is in contrast to the latter free to use.

The above image was generated by Clipdrop Stable Diffusion XL in response to a two-word text prompt, "hyperrealistic insanity."  The other three variations given by XL to this prompt were the expected closeups of faces so distorted by rage and anger their owners might very well have been rejected by Bedlam for having been too psychotic.  The above image was an anomaly.  It appears to show a devastated street in some war-torn country.  (Could it be Ukraine?)  I don't know what it was doing among the others unless the AI was far more perceptive than it is usually given credit for and was making a statement on the insanity of war.

At any rate, I decided to see what I would get if the image were "uncropped."  It has to be understood, of course, that in determining what should be put in the extended canvas of any image the AI must always be limited by the contents of the original.  Bearing that in mind, I thought the AI did a satisfactory job in generating the image shown below, although both the right and left sides of the extended image are lacking in detail.

I don't know how much use I will make of Uncrop in the future since it seems to offer an advantage only when, for one reason or another, a photographer needs a panoramic image in place of that actually shot.  If I do need a panoramic photo I would much prefer to simply shoot with my Sirui 50mm anamorphic lens rather than be bothered creating unreal images in post processing.

Tuesday, July 4, 2023

Picsart AI GIF Creator


The online photo editor Picsart has introduced a new AI feature, a GIF creator, that's at the moment at least free to use and that produces some truly offbeat imagery.  

As Picsart states on its website:
“A world of untapped GIF potential awaits at your fingertips. No longer will you have to rely solely on recycled content or endlessly search for the perfect GIF to convey your emotions or thoughts. Our AI GIF Generator empowers you to express yourself with unprecedented originality, opening up a realm of endless creativity. The GIF game will never be the same again.”
All one need do to use it is to enter a short text prompt.  It's not art, of course, but it might nevertheless prove useful to those looking to spice up their social media posts with unusual content.  It even offers a choice of styles to apply to the GIFs.  These include various moods, colors, and styles of famous artists.  I chose Steampunk for this example of a crowded NYC subway train moving down the track.

Monday, July 3, 2023

Text-to-Image AI Generated Prompts: ChatGPT vs. Google Bard

 

I recently came across a YouTube video in which its author compared images generated in Stable Diffusion XL with those generated in Midjourney.  There was nothing new there - such comparisons have been popping up all over the web during the past month.  What I found intriguing, however, was that the YouTuber used as prompts texts generated by ChatGPT based on only a single word.  I couldn't help wondering why the user had chosen GPT and not Google Bard which has lately been getting an equal amount of buzz.  I therefore decided to do a comparison of my own between the effectiveness of prompts generated by GPT from a single word as opposed to those generated in the same manner by Bard.  For this purpose, I limited myself to using SDXL as the sole image generator.

Going first to GPT, I asked the app to generate a prompt based on my own choice of word, in this case "beauty."  GPT immediately replied:
"Prompt: Create an AI-generated image that captures the essence of 'beauty' in nature.

Description: Imagine a serene landscape where beauty abounds. The image should showcase a harmonious blend of colors, shapes, and textures that epitomize the concept of beauty. Incorporate elements such as a stunning sunset or sunrise, vibrant flowers, majestic mountains, a serene lake, or any other natural elements that evoke a sense of awe and wonder. The goal is to convey the breathtaking beauty found in the natural world through an AI-generated image. Let your creativity and the power of AI bring this beauty to life!"
When I plugged this prompt into SDXL I got back the image shown below.  Just as GPT's prompt had been perfectly appropriate so was the image itself.  I thought it somewhat generic, though, and not particularly inspiring.


When I moved on to Bard I had a completely different experience.  Not only did the app provide five alternative prompts to the word "beauty" (each of them illustrated with watermarked images from, respectively, Open Art, Smithsonian, Unsplash, EarthSky, and Daily Hive) but before even beginning with those offered two unsolicited prompts of its own devising as follows:
"Sure, I can do that. Just give me a word and I will generate a prompt for you. For example, if you give me the word 'cat', I might generate a prompt like this:

A realistic 4k image of a cat sitting on a windowsill, looking out at the city skyline. The cat is wearing a red bow tie and has a content expression on its face."
"I can also generate prompts that are more abstract or creative. For example, if you give me the word 'dream', I might generate a prompt like this:

A surreal image of a dreamscape. The landscape is made up of strange and colorful shapes, and there are floating objects in the air. A figure is walking through the dreamscape, their face obscured by a mist."

I thought it pretty amazing that Bard could display such creativity.  As for the word "beauty," Bard suggested the following prompts:
"A realistic image of a field of flowers in full bloom. The flowers are a variety of colors, and they are swaying in the breeze. The sun is shining brightly, and the sky is a clear blue."
"A detailed image of a butterfly's wing. The wing is covered in intricate patterns, and it is reflecting the light in a beautiful way. The butterfly is perched on a flower, and its wings are slowly opening and closing."
"A stunning image of a sunset over the ocean. The sky is ablaze with color, and the clouds are reflecting the light in a magical way. The waves are crashing on the shore, and the seagulls are soaring overhead."
"A mesmerizing image of a nebula. The nebula is made up of gas and dust, and it is glowing with a beautiful light. The nebula is located in a distant galaxy, and it is millions of light-years away from Earth."
"A serene image of a mountain lake. The lake is surrounded by mountains, and the water is a clear blue color. The sun is shining brightly, and the trees are reflected in the water."

I was really floored by the versatility Bard showed in providing me with the above prompts and was in general very pleased with the images shown above that SDXL generated in response to them.  I intend to keep experimenting with Bard as a prompt generator and will continue to post here the images I obtain.