Nuvilab, established in 2018, is an award-winning “Food Vision AI” startup. It focuses on creating a singular mixture of {hardware}, software program, and cloud providers geared toward precisely measuring meals servings and waste in environments similar to cafeterias, eating places, buffets, and hospitals. Over the years, Nuvilab has amassed substantial information, enabling the delicate coaching of its AI software program, main to the newest AI Food Scanner 3.0.
Nuvilabs service depends on a number of scanning units that sense the sort and amount of meals served (or discarded) in varied locations. The aim is to precisely observe nutrient consumption, carbon emissions, and meals waste.
The information will be utilized to gauge meals consumption to create an correct report about meals consumption for an individual or group. Nuvilab avoids making “diet recommendations” as a result of extra context (private info) can be required. However, it may possibly rapidly floor info that might curiosity nutritionists or medical doctors.
The scanning units use two cameras. In lower than one second, they seize a stereoscopic view that produces picture and depth information. By combining each info, the AI can categorize the kind of meals within the picture and consider “how much” of it (= quantity) was served or discarded. From there, the software program can cross-reference a meals database and infer extra dietary particulars.
Different scanners are used relying on how the meals is served, for instance, cafeteria trays or buffet bins. That’s as a result of the required digital camera subject of view is completely different, and the buffet bin is a barely completely different use case because it tends to include a single sort of meals in substantial quantity.
There are some options to Nuvilab within the trade, however the ones I’ve heard of depend on scales to measure the amount of meals. That determination tends to be extra approximative but in addition has a unfavourable affect on the system deployment, which is why Nuvilab opted for a vision-based strategy. This additionally occurs to be ~50% inexpensive to deploy and keep, in accordance to the corporate.
Nuvilabs operates a cloud service for its prospects. Data from the scanners is distributed to a safe cloud (hosted on Amazon AWS) the place prospects can get insights from their information. New options added lately embrace an Inventory Manager to optimize purchases, however many different actions will be derived from the info.
The firm talked about working with Aramak in Korea. The similar firm can also be a big supplier of catering providers, together with right here in Silicon Valley. Having such companions may very well be nice for the service’s international enlargement as each firms can work carefully at each stage, together with updating the meals database since Amarak is aware of exactly what goes into the meals so far as I do know.
Nuvilab representatives talked about that the corporate is wanting on the hospitality market in North America, evaluating compliance wants within the European Union (essentially the most stringent worldwide), and believes that the Middle East will see essentially the most progress within the quick time period.
At CES 2024, Nuvilab will present its newest providers and reveal completely different options, together with diet consumption, kitchen optimizations, eating corridor use circumstances, and extra. You can discover our full dialog within the video beneath:
Filed in . Read extra about AI (Artificial Intelligence), CES, Food and Korea.