Fort Worth 24

collapse
Home / Daily News Analysis / I turned casual selfies into professional headshots with Gemini - and the results blew me away

I turned casual selfies into professional headshots with Gemini - and the results blew me away

Jul 09, 2026  Twila Rosenbaum  5 views
I turned casual selfies into professional headshots with Gemini - and the results blew me away

Professional headshots have traditionally required expensive photographers, studio lighting, and long sessions. But AI is changing that. Using Google's Gemini Nano Banana 2, you can transform casual selfies into polished, corporate-quality portraits in minutes. This article walks through the process and demonstrates results that can easily replace a costly photoshoot.

The prompt that works

The key to success is a well-structured prompt. The prompt used in this experiment is divided into three parts. First, it directs the AI to preserve the subject's facial features and hairstyle while altering the expression to convey confidence and authority. Second, it describes the lighting and background: soft studio illumination with delicate shadows, and an understated yet upscale backdrop—charcoal, deep taupe, or smooth black gradient. Finally, it captures the image in a premium editorial style with crisp focus and cinematic depth of field, as if shot on an 85mm lens. This combination consistently delivers headshots that look like they were taken in a high-end studio.

To achieve this, you can use the free tier of Gemini, though you may experience delays between generations. For faster access, the $20 Google AI Pro plan removes waiting times. The results are indistinguishable from traditional headshots, and with minor adjustments you can customize wardrobe, tie patterns, and even create fun variations.

Test case 1: Steven Vaughan-Nichols

Steven's original image already had a nice smile, but the goal was a formal corporate portrait. Using the three-part prompt, Gemini produced a hyper-realistic image with Steven in a dark navy suit, confident expression, and dramatic lighting. The AI maintained his exact facial features and hairstyle. The only change was a red regimental stripe tie, which was updated with a simple follow-up prompt: "Convert the tie to a red regimental stripe tie. Keep everything else identical and unchanged." The result was a perfect LinkedIn-worthy headshot.

Beyond corporate use, the same tool allows for creative fun. For Steven, a ZDNET Linux expert, a prompt requesting a "star performer of a metal band" with a T-shirt reading "Linux" and electricity shooting from a guitar produced an unforgettable publicity shot. This demonstrates the versatility of Gemini for both professional and whimsical portraits.

Test case 2: Artie Beaty

Artie provided three selfies, including one with his wife. The first step was to isolate him using a prompt that removed the woman and centered him facing forward. After that, the same three-part corporate prompt applied, resulting in a sharp headshot with perfect lighting. Once again, a red regimental tie was added with a simple instruction. The AI even introduced a classic crossed-arms pose, adding to the professional feel.

For a fun twist, Artie's interest in ghost tours led to a Ghostbuster-themed image. The prompt: "Dress the subject as if he's a Ghostbuster. Do not change his face or use facial makeup. Make the post appropriate for a publicity shot." The initial output incorrectly labeled him "Spengler," but a quick fix—"Remove the Ghostbuster patch and change the name to BEATY"—resulted in a uniform with perfectly rendered embroidered text. This showcases the AI's ability to handle text details accurately.

Why pay for dedicated headshot services?

Social media is filled with ads for AI headshot services charging significant fees. However, as this experiment demonstrates, you already have access to tools that can deliver the same—or better—results. With a careful prompt, you control every aspect of the image: lighting, style, wardrobe, and background. The quality rivals traditional photography, and the process is faster and more convenient.

The only minor drawback is the Gemini logo that appears in the corner of generated images. This can be removed with any photo-editing app using content-aware fill. But for many uses, the logo is unobtrusive. Overall, the ability to turn a casual selfie into a professional headshot without leaving home represents a significant advancement in AI image generation.

Expanding the experiment

To further test the limits, additional subjects could be tried. For instance, a group photo of colleagues could be turned into individual headshots by first extracting each person. The prompt structure remains the same: preserve facial features, control lighting and background, and specify editorial style. With practice, users can tailor results to match company branding, such as specific tie colors or background hues that align with corporate identity.

The technology behind Google's Gemini relies on large language models combined with diffusion image generation. The "Nano Banana 2" model, as it's called, is optimized for high-quality portrait output. Its ability to understand nuanced descriptions of clothing, expressions, and settings makes it a powerful tool for professionals who need polished photos without the expense of a photographer.

For those concerned about authenticity, AI-generated headshots can be easily detected by experts, but for everyday use on LinkedIn or company websites, they pass muster. The lighting and depth of field are often indistinguishable from real camera shots. As AI continues to improve, the gap will only narrow.

Practical tips for best results

Start with a high-resolution selfie. The AI works better with clear, well-lit source images. Avoid busy backgrounds; a plain wall works best. Write a prompt that explicitly states what to preserve and what to change. Use the three-part structure: subject description, lighting/background description, and camera style. Iterate if needed—regenerate a few times to get the ideal expression or lighting. The AI's randomness can produce varied results, so patience pays off.

Additionally, you can combine multiple prompts. For example, generate a corporate headshot first, then adjust the tie or add a fun element. The AI retains the same subject identity as long as you refer to the previous image. This allows for a library of variations from a single source photo.

For teams, a single selfie can be turned into consistent headshots for everyone. Simply apply the same prompt to each person's photo, ensuring a uniform style across the organization. This is far cheaper than hiring a photographer to shoot each employee individually.


Source: ZDNET News


Share:

Your experience on this site will be improved by allowing cookies Cookie Policy