Introducing a system that divides the purchase price by the number of times the item has been worn and calculates the ``price'' of the outfit for that day. Improve accuracy and efficiency with Gemini.
AI-generated summary
An article introducing the system that Takako Kantai has been operating at her home for nearly six years and exhibited at Maker Faire Tokyo 2026.
ITmedia NEWS Weekly AccessTop10
AI automatically determines “the clothes you wore today” The system that shows the “cost performance of clothes” for the day is amazing
ITmedia NEWS Weekly AccessTop10
August 29th - September 4th, 2026
Last week's most visited item was a ``supercar bicycle'' for adults that is equipped with retractable lights and RFID authentication. Isn't it a contradiction to say "bicycle" when it's "car"? That's what I thought, but it turns out it's a reproduction of a bicycle that boys dreamed of in the 1970s, equipped with supercar-like equipment.
Now, although this has nothing to do with rankings, I helped exhibit an interesting project called ``visualizing the cost performance of clothes,'' so I would like to introduce it to you.
A system that shows the “price” of the day’s outfit.
Even if you buy a shirt for 20,000 yen, if you wear it 100 times, it will cost you 200 yen per day. Shoes that cost 50,000 yen cost 100 yen for the 500th time. On the other hand, the shorts I bought yesterday for 3,000 yen and are wearing for the first time cost 3,000 yen because it's my first time wearing them. The total cost of today's outfit is 3,300 yen.
This is the data that is provided to me every day by a system created by my friend Takako Kantai. Divide the purchase price by the number of times you wear it and calculate the "price" of the outfit for that day. She continues to use it for herself at home and continues to brush it (from Takako Kanzai's X).
Recorded over 2,000 times for nearly 6 years...Improved efficiency through the use of AI
The outfits are recorded by a camera equipped with a Raspberry Pi + sensor placed at the entrance of Kantai's home. Take a full body photo with this camera just once before going out.
The clothes you own are registered in advance along with the price at the time of purchase, and Gemini compares them with a full-body photo and determines that ``I'm wearing this top and these pants today'' and stores them.
In the past, they used dedicated apps or built their own wardrobe judgment tools, but these were time-consuming and the judgment accuracy was questionable. Since Gemini has started to rely on Gemini to judge clothes, it seems that the accuracy and efficiency have improved significantly.
The cost of each item decreases each time you wear it. When you accumulate this information, you can see at a glance the daily cost of clothing, the cost performance of each item of clothing, and which items you often wear and which items you don't wear at all.
Mr. Kantoshi has been using this record for nearly six years, changing the system, and has succeeded in visualizing ``clothes that he buys but never wears'' and getting rid of. Currently, she only owns 20 tops and 8 bottoms all year round. It's unbelievably small for a woman's wardrobe.
Women "want it", men "don't want it but it's interesting"
Mr. Kantai exhibited this system at the event "Maker Faire Tokyo 2026" held this weekend, and I helped out.
Many people at the venue were interested in the system, and expressed their admiration for how the system works, the labor-saving benefits of AI, the fact that clothes can be "depreciated," and the fact that Mr. Kantai has been keeping records for nearly six years.
On the other hand, different trends were observed between men and women. Women often said, ``I want it,'' and men often said, ``It's interesting, but I don't need it because I always wear the same clothes.''
The source code for this system is publicly available on GitHub, and anyone can copy and create it. One woman said, ``I'll have my husband make one for me,'' and another man said, ``My girlfriend made a similar system for me,'' which was very impressive.
I want it too! That's what I thought, but even when I looked at the source code, it was pretty clear, and I had so many clothes that I didn't feel like I could pull them all out and figure them out. I definitely have a lot of clothes that I don't wear...

In his keynote speech at Hitachi Social Innovation Forum 2026 JAPAN, Hitachi CEO Toshiaki Tokunaga announced the expansion of HMAX, a social infrastructure solution using physical AI. We will develop new services while simultaneously reducing costs and ensuring safety.

Razer has announced that it will release the Clio X head cushion with wireless speakers in Japan in September. The price is 17,980 yen. Additionally, a gaming keyboard in collaboration with the e-sports team "Sentinels" will be released.

OpenAI has strengthened safety measures for its new AI "Astra" and announced that the rejection rate of cyber requests has increased to 91.5%. In light of the ``AI escape incident'' that occurred this summer, while mechanisms such as real-time monitoring have been introduced, concerns have also been raised about ``deception'' in which AI follows rules only during testing.

ACEMAGIC announced the AI mini workstation "F9A" equipped with AMD's latest processor "Ryzen AI Max+ PRO 495" at IFA 2026 in Germany. Equipped with up to 192GB of memory, it also supports local execution of large language models.

Gartner has positioned physical AI and agent-based AI at the peak of "excessive expectations" in the 2026 Hype Cycle. Physical AI will become the basis for restructuring entire industries, including manufacturing, around data, and agent-based AI is moving toward improving operational efficiency and creating new business models by automating some of the complex tasks. Generative AI-related technologies are expected to be widespread in less than two years.

In response to the increasing number of 110 calls received, the National Police Agency has decided to develop an AI system that automatically converts call audio into text and analyzes the level of urgency. The aim is to quickly and accurately record the contents of reports and eliminate variations in interviews by responding police officers.