Nomitri-Background

About Us

We are a deep-tech AI Start-up located right at the heart of Berlin.

As a team, we have decades of work experiences and in-depth know-how in the field of artificial intelligence and business operations.

Coming from the autonomous vehicles industry and medical imaging research, we bring a unique combination of software and hardware development expertise to take deep-learning visual perception AI to the next level.

About Us

We are a Berlin-based deep-tech AI Start-up located right at the heart of Berlin.

As a team, we have decades of work experiences and in-depth know-how in the field of artificial intelligence and business operations.

Coming from the autonomous vehicles industry and medical imaging research, we bring a unique combination of software and hardware development expertise to take deep-learning visual perception AI to the next level.

Our Vision

Visual perception AI in your pocket, everywhere and offline!

Imagine every mobile phone gets the intelligence to „recognize“ objects and actions like a human being enabling a whole host of new applications that empower its users to interact with the world around them in so far unimaginable ways. 

And all of that in a privacy preserving way, without any sensor data ever leaving your device. 

Our Vision

Visual perception AI in your pocket, everywhere and offline!

Imagine every mobile phone gets the intelligence to ‚recognize‘ objects and actions like a human being enabling a whole host of new applications that empower its users to interact with the world around them in so far unimaginable ways.

And all of that in a privacy preserving way, without any sensor data ever leaving your device.

The Challenges
We Tackle

In order to bring deep learning visual perception AI to edge devices in a commercially viable manner, we have to enable our neural network based software to run in

  • limited processor and memory environments
  • with limited energy supply
  • in realtime
  • without loosing significantly in accuracy

The Challenges
We Tackle

In order to bring deep learning visual perception AI to edge devices in a commercially viable manner, we have to enable our neural network based software to run in

 

  • limited processor and memory environments
  • with limited energy supply
  • in realtime
  • without loosing significantly in accuracy

Our Solution

Deep-learning visual perception solutions that are well integrated with the hardware components, tailored to their use-cases and that make use of the latest advances in network pruning and quantization techniques.

We use a variety of techniques that enable us to greatly reduce the memory, runtime and energy requirements of deep learning based image recognition (by more than 90%) without significantly sacrificing accuracy (~1 to 3%).


At the same time, our approach allows us to minimize the effort needed for annotating training data for our deep learning algorithms, which in turn makes new commercial applications profitable and thus commercially viable.

Our Solution

Deep-learning visual perception solutions that are well integrated with the hardware components, tailored to their use-cases and that make use of the latest advances in network pruning and quantization techniques.

 

We use a variety of techniques that enable us to greatly reduce the memory, runtime and energy requirements of deep learning based image recognition (by more than 90%) without significantly sacrificing accuracy (~1 to 3%).


At the same time, our approach allows us to minimize the effort needed for annotating training data for our deep learning algorithms, which in turn makes new commercial applications profitable and thus commercially viable.

Potential Use Cases

Potential Use Cases

Government

We can help governments increase public security through camera monitoring – BUT in a privacy compliant way:

  • cameras are able to recognize dangerous behavior patterns in the device itself
  • no need to store vast amount of video data for later analysis
  • only suspicious behavior triggers temporal video storing, all other material will never get stored
  • human (e.g. judge) then decides if there is a case to allow for permanent storage in police archives

Industry

We can help industrial edge devices to become smart in recognizing and classifying objects that they view and send results of analytics (not raw image data) back to headquarter for action or insight consolidation.
This results in significantly less bandwidth & storage requirements and longer operating times:

  • smart drones monitoring for maintenance
  • production process monitoring
  • agricultural monitoring
  • weather monitoring

Consumer

Your deep learning enhanced mobile phone expands your ability to recognize and analyze whatever your see through your mobile camera lense, such as 

  • basic visual medical checks (e.g. skin moles)
  • used vehicle purchase checks 
  • lie detector (private purchases, dating, …) 
  • sightseeing information
  • additional product informations (vegan, allergies)

Government

We can help governments increase public security through camera monitoring – BUT in a privacy compliant way:

  • cameras are able to recognize dangerous behavior patterns in the device itself
  • no need to store vast amount of video data for later analysis
  • only suspicious behavior triggers temporal video storing, all other material will never get stored
  • human (e.g. judge) then decides if there is a case to allow for permanent storage in police archives

Industry

We can help industrial edge devices to become smart in recognizing and classifying objects that they view and send results of analytics (not raw image data) back to headquarter for action or insight consolidation.
This results in significantly less bandwidth & storage requirements and longer operating times:

  • smart drones monitoring for maintenance
  • production process monitoring
  • agricultural monitoring
  • weather monitoring

Consumer

Your deep learning enhanced mobile phone expands your ability to recognize and analyze whatever your see through your mobile camera lense, such as 

  • basic visual medical checks (e.g. skin moles)
  • used vehicle purchase checks 
  • lie detector (private purchases, dating, …) 
  • sightseeing information
  • additional product informations (vegan, allergies)

Want to work at the forefront of Deep Learning innovation? We are hiring!