In Pursuit of Smart City Vision, Startup Two-i Keeps an AI on Worker Safety

” It would be like having a weather map of the city, only one that measures performance,” said Trombini, who serves as chairman of Two-i, an NVIDIA Metropolis partner based in Metz, a historical city in northeast France.

” The algorithms are exactly the same as what were utilizing for different clients,” said Trombini. “Its the same technology, however its packaged in a various method.”

Trombini and Cazenave imagined a system that would enable civic leaders to see what improvements were required across a municipality.

Two-i creators Julien Trombini (left) and Guillaume Cazenave.

While this use case is extremely specific, the businesss AI architecture is developed to flexibly support various algorithms and functions.

Together with the energy giant, Two-i has produced an AI-enabled video analytics application to identify when individuals near a risk zone and threat falling and instantly alert others to take quick action. In its preliminary months of operation, the vision AI system avoided 2 mishaps from taking place.

When Julien Trombini and Guillaume Cazenave founded video-analytics start-up Two-i four years back, they had an enthusiastic objective: enhancing the quality of urban life by one day having the ability to keep track of a citys roadways, garbage collection and other public services.

Along the method, the pair discovered a completely various niche. Today, the businesss innovation– which combines computer system vision, information science and deep knowing– is assisting to prevent lethal accidents in the oil and gas industry, among the worlds most unsafe sectors.

That showed a high order, so the two refocused on specific facilities, such as stadiums, retirement homes and transit stations, where its tech aids with security and event detection. It can signal the right individuals when a retirement home citizen falls in a passage. Because of a damaged lift, or when a transit rider utilizing a wheelchair cant get on a train.

More just recently, the company was approached by ExxonMobil to aid with a possibly fatal concern: improving worker safety around open oil tanks.

Taking advantage of Vision AI

” The time we can save there is essential, and the better the hardware, the more precise the results and faster we get to market,” said Trombini.

Two-i taps the CUDA 11.1 toolkit and cuDNN 8.1 library to support its deep knowing procedure, and NVIDIA TensorRT to speed up inference throughput.

The business aims to often upgrade its GPUs to guarantee its offering clients the fastest possible option, no matter how lots of cams are feeding data into its system.

Two-i has actually likewise begun benchmarking NVIDIAs pretrained designs against its algorithms and started utilizing the NVIDIA DeepStream SDK to enhance its video analytics pipeline.

” The much heavier the algorithm, the more expensive,” he stated. “We utilize the TAO toolkit to prune algorithms and make them more tailored to the task.”

Two-is versatility originates from its reliance on using the NVIDIA Metropolis platform for AI-enabled video analytics applications, leveraging innovative tools and adopting a full-stack method.

Training that initially took up to two weeks has actually been slashed to 3 days utilizing the NVIDIA TAO Toolkit, a CLI- and Jupyter Notebook-based version of the NVDIA train, adapt and optimize framework.

Trombini states one of the most engaging pieces of NVIDIA tech is the NVIDIA TAO Toolkit, which helps the business keep costs down as it tinkers with its algorithms.

To do so, it counts on a variety of NVIDIA-Certified Systems, using the latest workstation and information center GPUs based on the high-performance NVIDIA Ampere architecture, for both training and reasoning. To reduce training times further, Two-i is wanting to evaluate its substantial image dataset on the effective NVIDIA A100 GPU.

Building on Success

Two-i sees its ability to fix complex issues in a variety of settings, such as for ExxonMobil, as a springboard to swinging back around to its original clever city aspirations.

Already, its keeping an eye on all roadways in eight European cities, evaluating traffic circulations and comprehending where vehicles are coming from and going to.

” Its coming slowly,” he stated, “but we are starting to execute our vision.”

Trombini recognizes that Two-i has to keep its focus on providing one advantage after another to accomplish the businesss long-term goals.

Leave a Reply

Your email address will not be published. Required fields are marked *