您现在的位置是:New AI model suggests recipes based on available food ingredients >>正文
New AI model suggests recipes based on available food ingredients
上海品茶网 - 夜上海最新论坛社区 - 上海千花论坛4973人已围观
简介By subscribing, you agree to our Terms of Use and Policies You may unsubscribe at any time.Artificia...
By subscribing, you agree to our Terms of Use and Policies You may unsubscribe at any time.
Artificial intelligence (AI) has dabbled into many things, from poetry to video creation. And now, it seems that researchers using AI are cooking up something new.
Researchers at PeopleTec, an Alabama-based technology company, have created an AI-based model that can create recipes based on the ingredients provided by users.
"The basic idea behind our work was to combine raw food and recipe ingredients using image analysis, then to ask a powerful language model to construct a plausible cooking recipe, including the expected title, proportion, and steps," David Noever, one of the researchers, told TechXplore.
According to the study, this computational model can also simply scan your refrigerator and recommend a recipe for your next meal. “To demonstrate the API as a modular alternative, we solve the problem of a user taking a picture of ingredients available in a refrigerator, then generating novel recipe cards tailored to complex constraints on cost, preparation time, dietary restrictions, portion sizes, and multiple meal plans,” researchers wrote in the paper uploaded on a pre-print server.
See AlsoThe training of the AI model
The computational model was trained using over 2,000 images of open refrigerators containing various raw food items. As a result, the AI model was able to create a 100-page recipe book consisting of various unique recipes using these image inputs.
“For the first time, an AI chef or cook seems not only possible but offers some enhanced capabilities to augment human recipe libraries in pragmatic ways. The work generates a 100-page recipe book featuring the thirty top ingredients using over 2000 refrigerator images as initializing lists.” the paper noted.
The AI method is based on models that can recognize objects in images along with a text generator model to give out recipes. Specifically, they used application programming interfaces (APIs) and GPT-4, an OpenAI large language model (LLM), to generate recipes based on text or images provided by the users.
The researchers hope to incorporate this AI model into a smartphone app or other software tools to generate new and innovative recipes.
Study abstract:
The AI community has embraced multi-sensory or multi-modal approaches to advance this generation of AI models to resemble expected intelligent understanding. Combining language and imagery represents a familiar method for specific tasks like image captioning or generation from descriptions. This paper compares these monolithic approaches to a lightweight and specialized method based on employing image models to label objects, then serially submitting this resulting object list to a large language model (LLM). This use of multiple Application Programming Interfaces (APIs) enables better than 95% mean average precision for correct object lists, which serve as input to the latest Open AI text generator (GPT-4). To demonstrate the API as a modular alternative, we solve the problem of a user taking a picture of ingredients available in a refrigerator, then generating novel recipe cards tailored to complex constraints on cost, preparation time, dietary restrictions, portion sizes, and multiple meal plans. The research concludes that monolithic multimodal models currently lack the coherent memory to maintain context and format for this task and that until recently, the language models like GPT-2/3 struggled to format similar problems without degenerating into repetitive or non-sensical combinations of ingredients. For the first time, an AI chef or cook seems not only possible but offers some enhanced capabilities to augment human recipe libraries in pragmatic ways. The work generates a 100-page recipe book featuring the thirty top ingredients using over 2000 refrigerator images as initializing lists.
Tags:
转载:欢迎各位朋友分享到网络,但转载请说明文章出处“上海品茶网 - 夜上海最新论坛社区 - 上海千花论坛”。http://www.jz08.com.cn/news/65153.html
相关文章
Bitcoin price prediction for June — is the microbear market nearly over?
New AI model suggests recipes based on available food ingredientsWith summer finally here, can the Bitcoin market finally start to heat up again?Since hitting an all...
阅读更多
Where to buy AquaGoat as price rallies by 25%
New AI model suggests recipes based on available food ingredientsAquaGoat’s price is up by 25% over the past 24 hours and could rally higher due to massive int...
阅读更多
Where to buy Marsan Exchange Token as MRS sees 207% rally
New AI model suggests recipes based on available food ingredientsThe Marsan Exchange token (MRS) is up by 207% and is trading at $0.08 at the time of writingThe Mars...
阅读更多
热门文章
- Top 4: Cryptocurrencies to watch in the last quarter of 2020
- The best places to buy DOGET coin: Doge Token surges in value
- Digital Currency Not On The Cards For New Zealand Right Now
- Where to buy HakunaMatata (TATA)
- Financial Bodies Lag Behind in Tracking Crypto Transactions
- Beam wallet brings Amazon and Shopify purchases to users