AI Product Photography Without a Studio: The VdoBloom Workflow

You can shoot studio-grade product photography without a studio by photographing the product once on a plain surface with a phone, then using AI image editing to replace the background, fix the lighting and generate every angle and lifestyle variant you need — on VdoBloom a finished marketplace-ready image costs between 2 and 19 credits, which is $0.10 to $0.95. The part that still requires you is the single honest source photo. Everything downstream — the seamless white sweep, the marble pedestal, the sunset kitchen counter, the 4K hero crop — is a prompt.

What AI can and cannot replace

Be clear about the boundary, because this is where most people waste credits. AI editing models are extremely good at re-contextualising a product: new background, new surface, new light direction, new colour grade, added props, extended canvas. They are much weaker at inventing a product they have never seen. If you feed a blurry, badly lit photo where the label is unreadable, no model will hallucinate your label correctly. Text on packaging, fine engraving, stitch lines and logo kerning all come from your input pixels.

So the rule is simple: spend your effort on capture, not on lighting. A phone, a window, a sheet of white paper as a bounce card, and a tripod or a stack of books will get you a sharp, evenly exposed, in-focus frame. That is enough. Shoot it square-on and shoot one three-quarter angle. Those two frames feed an entire catalogue.

The studio-free workflow, step by step

1. Capture flat and sharp. Daylight from a window, product on a clean surface, no direct sun, no on-camera flash. Fill the frame. Do not apply phone filters or beauty modes.
2. Cut it out and rebuild the set. Open the AI image editor, upload the photo, and describe the scene you want: seamless white cyclorama, soft overhead key, a soft contact shadow directly beneath the product. Name the shadow explicitly — products floating without a shadow are the single most obvious AI tell.
3. Clean the defects. Dust, a stray reflection of your hand, a price sticker. Use the erase model on that region rather than re-prompting the whole image, so the rest of the frame stays byte-identical.
4. Extend for every placement. One 1:1 listing image, one 16:9 banner, one 9:16 story. Outpainting expands the canvas outward and matches the existing light instead of stretching pixels.
5. Finish at the resolution the channel wants. Marketplaces generally want a large, clean file. Re-run the winning prompt at 2K or 4K rather than upscaling a 1K result.
6. Add motion if the channel rewards it. A turntable spin from the same still often outperforms a static image on a product page. The Product 360 Spin effect takes the exact photo you just finished.

What each step actually costs

VdoBloom prices per generation in credits, and 1 credit is $0.05 on the Lite plan ($15 for 300 credits). These are the real numbers from the credit table, not estimates.

Job	Model to use	Credits	Cost	What you get
Fast draft / concept	Z-image Turbo	2	$0.10	Cheapest look-test before you commit
Background swap	Nano Banana Edit	5	$0.25	Reliable everyday product edit
Multi-reference edit	P-Image-Edit	6	$0.30	Combine product plus a background reference
Precise instruction edit	Flux Kontext Pro	7	$0.35	Follows single-change instructions tightly
Listing image at 1K	GPT Image 2 image-to-image	3	$0.15	Cheapest route to a clean square listing shot
Listing image at 2K	GPT Image 2 image-to-image	5	$0.25	The usual marketplace sweet spot
Hero image at 4K	GPT Image 2 image-to-image	8	$0.40	Banner and print-safe resolution
Remove an object	FLUX Erase	10	$0.50	Targeted cleanup, rest of frame untouched
Extend the canvas	FLUX Outpainting	12	$0.60	Reformat 1:1 to 16:9 without cropping
Premium 2K render	Nano Banana Pro	18	$0.90	Top-tier detail and text handling
Flagship finish	Seedream 5.0 Pro (High)	19	$0.95	Highest-fidelity edit tier
360 spin video, 720p 5s	Seedance 2 Mini	17	$0.85	Turntable clip from the finished still

A realistic first product: one draft at 2 credits, three background attempts at 5 credits each, one erase pass at 10, one outpaint at 12, one 2K final at 5. That is 44 credits, $2.20, for a full listing set. New accounts start with 10 free credits and no card, so the first draft cycle costs nothing.

Prompting for product shots specifically

Product prompts fail differently from portrait prompts. The model does not need to be told the product is beautiful; it needs to be told what not to touch. Three habits fix most output:

• Lock the subject in words. Write that the product stays identical to the original photo, with labels, logos and text sharp and legible, and no warping or resizing. Models respond to explicit preservation language.
• Describe light physically, not emotionally. “Soft diffused overhead key, gentle fill from camera left, soft contact shadow beneath” beats “professional studio lighting.”
• Change one thing per generation. If background and angle and colour all change at once and the result is wrong, you cannot tell which instruction caused it. Chain edits and review each step.

For lifestyle shots — the mug on a breakfast table, the serum on a bathroom shelf — name the surface material, the time of day and the depth of field. Shallow depth of field is what separates a lifestyle image from a composite.

When to generate instead of edit

If you do not have the product yet — a pre-order, a concept, a dropship item you only have a supplier render for — editing has nothing to preserve. Use the text-to-image generator to build the scene and treat the output as a mock-up, not a listing image. Selling a generated image as a photograph of a product a customer will receive is a returns problem and, on most marketplaces, a policy problem.

The honest split: edit when the product exists, generate when it does not, and never present a generated concept as the item in the box.

Where the intent pages help

Several ready-made routes wrap this workflow with defaults already set. The AI product photography tool is the general entry point; white-background product video targets marketplace listing requirements specifically. Pricing for every plan and the one-time credit packs that never expire is on the pricing page — packs start at $2.49 for 75 credits, which is roughly three finished listing images.

Paid plans download watermark-free with commercial rights, which matters here: a watermark in the corner of a listing photo is not usable, so a product catalogue is one of the few use cases where the free tier is genuinely a trial rather than a workflow.

Frequently asked questions

Do I need a lightbox or a photo tent?

No. A window, a white wall or sheet of paper, and a steady phone give a clean enough source frame. The background, sweep and lighting are all replaced in editing anyway, so money spent on a tent is money that would be better spent on credits.

Will the AI keep my label text readable?

Only if it is readable in your input. Models preserve text far better than they invent it. Shoot the label in focus and square-on, and state in the prompt that all text must stay sharp and legible. If a label still degrades, try a higher tier such as Nano Banana Pro at 18 credits, which handles typography noticeably better.

How many credits should I budget per product?

Around 40 to 60 credits per product for a full set including retries, so $2 to $3. A 300-credit Lite month covers roughly five to seven products end to end, and unused one-time pack credits never expire if your volume is lumpy.

Can I use these images on Amazon, Shopify or Etsy?

Yes, on a paid plan, which includes commercial rights and watermark-free downloads. Check each marketplace’s own image rules separately — some require a pure white background and a minimum pixel dimension, which is why the 2K and 4K tiers exist.

Is a spin video worth the extra credits?

On a product detail page, usually yes. A 720p five-second turntable from Seedance 2 Mini is 17 credits, $0.85, and reuses the still you already paid for. Shoot once, edit once, animate once.