Flair.ai Review: How Useful Are its AI Tools for Product Photos?

There are some aspects of photography that are, frankly, a bit of a hassle.

One of those is trying out various studio settings for product photography. As a contract photographer, I’ve pretty much seen every possible attempt at making product photos look cool.

And usually, that boils down to three main components: having a studio with excellent lighting, a high-quality camera, and endless patience to shoot and reshoot the same thing over and over again.

The main issue is that product photos typically only work well when they’re cohesive with the manufacturer’s (or designer’s) branding, which means that you might need to go through a lot of iterations.

Enter AI tools, which promise to revolutionize the entire industry. (Case in point: these 6 AI tools revolutionizing e-commerce photography.)

While I’m always hesitant about using AI generators to replace my entire workflow, they can be extremely efficient in repetitive tasks or finding and sticking to patterns.

Flair.ai is a content creation platform that aims to take many of those time-consuming jobs off your plate. Its main draw is being a one-stop shop for product photography. So let me take you through its features to see what you can do with it.

What Is Flair.ai?

Think of Flair.ai as a full-service photo editing studio right in your browser. It’s designed to be fairly simple to use while leveraging some of the most advanced image-generation models on the market.

The main draw of the tool is that it’s entirely browser-based.

While I love a good app myself (and the ability to snap and edit on the go), using a browser means you’re likely to have fewer issues working around odd features or a UI designed for a small screen.

The AI’s suite of tools is equally impressive, and all have to do with product photography in one way or another.

Feature Review

Background Remover and Replacer

One of the first things that any AI does when you upload your image to edit it is remove the background.

For Flair, I haven’t noticed many discrepancies when it comes to background removal.

There’s the occasional pixel or two of an outline where the AI couldn’t exactly delineate between the background and the product, but overall, it’s quite solid.

Notably, the background remover isn’t a separate tool per se. Instead, it will only be available when you upload an image, with the AI prompting you for the next step.

Now, as far as background generation is concerned, I always like seeing a large template gallery, and Flair.ai is really useful in this regard.

As soon as you upload an image, you get a grid with more than a dozen options and a textbox if you want to generate a background based on a custom text input.

I’ve tested the template backgrounds, and they’re mostly OK-ish. The tool is essentially pulling from that stock background and just placing your product image in front.

From the image above, you can see how the AI prompts you to place the product relative to the background so it can be filled in.

However, once it does get the template, the result is actually much better. The tool doesn’t just place the product; it also applies custom shadow and lighting corrections to make sure it fits the intended “studio background.”

The custom background tool is useful (and can generally understand complex prompts). However, for most purposes, if you’re working with AI, the templates will likely be enough when starting.

Notably, you do get an odd result or two, where the lines don’t seem to match, or background items appear from apparently nowhere. So, the “Improve Result” button might be necessary if you want to get an image that’s a real keeper.

Product Image Generation

Beyond just replacing the background, the tool’s primary function is to create a detailed scene surrounding the product, including a human model that will wear or hold it.

First, I used the generator without a human model, focusing only on the product (meaning that I didn’t put instructions on where a hypothetical model would be placed or how it would be posed).

The two images that I got for each test were usually similar enough to be interchangeable, and all matched the intended result from the textbox.

However, some, again, had odd bits and pieces that would require a regeneration.

For example, the tool doesn’t exactly recognize what it’s trying to generate for, so it put in lime slices as if I were making a martini commercial.

Video Generation

For video generation, the suite can start from where you left off in terms of scene generation, directly inputting the exact prompt you used to be expanded upon.

I could also choose from eight stock camera movements (pretty much covering all basic directions or rotations).

However, do note that even though the pricing and data usage tabs suggest that you get one free video per day, as soon as you input a caption or camera instruction, you’ll be blocked from generating one.

Obviously, video creation uses up a lot of AI resources, and it makes sense that it would be locked behind a paywall. But it doesn’t really make sense to advertise something that doesn’t actually exist in the free version to begin with.

As far as the video quality is concerned, the tool is alright with basic motions, but it can get confused if you try to move objects that are not parts of the product or the human model that the AI generated.

Items in the background can sway weirdly, and my prompt to have one lime roll required multiple generations to get a single usable result.

Drag-and-Drop Canvas

One of the more interesting (and, sadly, less intuitive) uses of the tool is the drag-and-drop canvas.

In theory, the canvas allows you to select your product from the gallery, then populate the background and foreground with other objects that will be slightly changed in the final result.

But here’s where you might start running into problems because the tool is arguably a bit overengineered.

Since the background remover tool doesn’t exist in isolation, you can’t simply upload a product image to use with the template.

Instead, the tool will select the original image and everything else, which will invariably be used and can “corrupt” the result.

The tool also doesn’t provide a useful “in-progress” bar. Instead, it shows a “real-time” processor that sometimes looks like it’s pulling from the canvas and generating an image.

This doesn’t actually do much, but the “Generate” button prompts the final image generation (and is more reliable). I used the “blend to scene” option with the original image, and all I got was a transparent box where the previous background used to be.

What the tool does have is an extensive props gallery, ranging from platforms to backgrounds to random objects that you can scatter for scenery.

Overall, this can be a solid tool, but you’d need to find a way to remove the background first and save the result to Flair so it can be usable.

AI Model Builder for Products

In general, whenever you want to create an image using Flair, the tool will ask you to upload a few images from different angles. For fashion and jewelry, this might require the use of a human to model them.

The images in question don’t have to be full studio quality, but you still need to make sure that the image resolution is high enough and that there isn’t a lot of background noise.

Then, the tool can take quite a bit of time to post the final result, complete with a description of the product that will be used for future generations and a link to the exact model in your library.

There’s an option for “Instant” generation, but this is reserved for the higher subscription tiers. As such, if you have a lot of product to go through, expect this to take a bit of time.

AI Human Model Generator

Beyond creating models of products, the AI can also generate a human model that can be used for your advertising.

This is perhaps one of the most polarizing features in my line of work.

On one hand, it provides a low-cost, lower-quality solution to hiring and scheduling both a photographer and a human model (while protecting your privacy).

On the other hand, humans in the industry are being slowly replaced by a litany of AI images, which creates an unfortunate precedent that others might follow.

But ethical issues aside, the model builder comes with a surprising range of options. You can customize the model’s age (or rather stage of life), race, gender, skin tone, facial features, hair, and even body type up to and including breast size.

Of course, it goes without saying that this is reserved for advanced subscriptions, and getting more than one model requires you to go up a tier.

However, once you do finish creating your model human, it becomes available for drag-and-drop canvases, image generation, and video creation, so it’s one of the more worthwhile designs that justify the time and money investment.

AI Image Editor Canvas

Beyond straight-up scene generation, Flair also has a function-based AI editor that can perform a few basic AI edits on an existing image.

In here, you’ll find some of the more useful features, such as image quality improvement, sizing changes, product replacement, and even human model changes.

However, the entire suite is only reserved for the Pro+ subscription tier, which can be prohibitively expensive considering what you get in return.

As such, I’d only recommend subscribing to that tier if you want the advanced model generation and this as a bonus rather than the focal point of the subscription.

Subscriptions and Cost

Unlike most AI suites, Flair.ai offers five free generated images and one video generation to test the toolset out, plus what is supposed to be three minutes of drag-and-drop canvas editing.

The pricing page mentions a free video, and the “data usage” tab indicates you should be able to train a small model for free, but both of these features seem to have been removed and redirect to the subscription page.

In terms of cost, Flair.ai’s basic subscription tier costs US$10 per month if you want to stick with a monthly subscription to try out the various tools over the first month.

However, that basic subscription doesn’t actually offer all that much, giving you a video generation and a bit more time using the drag-and-drop canvas.

The main Pro+ subscription gives you access to advanced models, model training, and the AI image editor, plus removes most of the bottlenecks in terms of image and video generation credits.

However, considering that Pro+ costs a significant US$26 per month and doesn’t actually have a free trial to test out these tools in earnest, your only option is to dive in and make the most out of the first 30 days of testing to check if the tool is really right for you.

Who Is Flair.ai For

As per usual, this AI suite is unlikely to replace a human photographer any time soon.

In fact, from what I’ve seen during testing, the tool’s biggest weakness is that it requires images to be relatively high-quality to begin with in order to improve them.

With this in mind, the tool is primarily useful for entrepreneurs or small businesses who only have a few new products to showcase on their social media and e-commerce pages per month. At these volumes, the tool’s AI generator makes it a good budget replacement for a human model.

At higher tiers, you do get a bit of savings, which is where the ability to create a human model from scratch can come in handy. It essentially allows you to reduce the cost of photoshoots while having a standardized image quality.

Ultimately, what I liked about the tool is that it has a lot to offer, even if it’s a bit pricy.

What I didn’t particularly like is that the creators have this sneaky approach where they directly advertise services that aren’t actually there in some plans (there’s even an “affiliate” option at the top of the editor that also leads nowhere).

This makes me lose just a little bit of trust in the product’s longevity and could lead to features straight up getting redelegated to higher subscription tiers out of nowhere.

So, if you do decide to dive in and use the tool, just beware that nothing seems to be set in stone.