Persival GmbH’s Post

Captain AI and VSTEP Simulation are revolutionizing Maritime Machine Vision by leveraging the Nautis simulator alongside Generative AI to generate infinite datasets for Machine Vision AI models! This collaboration combines VSTEP's state-of-the-art simulation capabilities with Captain AI's expertise in autonomous navigation and situational awareness, overcoming the traditional data collection methods' limitations such as stereo cameras and LiDAR. While these methods are effective, they often incur high costs, complex setups, and fail to fully capture critical edge cases. This breakthrough is a significantly more cost-effective and scalable alternative to real world data collection. “We're pioneering a method for generating accurate 3D spatial data for single-view cameras” said the CTO Gerard Kruisheer “setting a new benchmark for efficiency and simplicity in data collection”. This advancement will underpin Captain AI’s cutting-edge 3D bounding box detection model which is currently uder development and promises to be future foundation of maritime safety and operational intelligence.

Leveraging Nautis simulator with generative AI to get one step closer to the singularity 🚢

Its exciting to watch Terra AI from Stereolabs take shape on their website! #SpatialAI is finally mature enough for widespread adoption by autonomous robots and they're leading the charge powered by NVIDIA Robotics Jetson boards. As expert roboticists, we recognize the unique challenges inherent with deploying and managing modern AI-based perception systems (doing far more than just 2D detection and segmentation), so having partners like Stereolabs is critical to accelerating your #ROS, #Nav2, or other product to market. I can't wait for a future where Nav2 and autonomy frameworks are managing the standardized outputs of dense and semantic environmental models from #TerraAI and others for our planning, control, behaviors, or even higher level #AI application systems! There will be so much more we can do to make robots safer, more reliable, and more intelligent. Definitely check out the great videos they have there!

🤖 Unlocking Robots' Vision: Their Perspective on the World! 🌍 Have you ever wondered how robots see the world around them? There is more to robot vision than simply cameras and sensors, therefore let's explore this intriguing field: 1️⃣ Lidar Technology Robots employ lidar, a technology that maps the surroundings in three dimensions using lasers. It is not without limits, though, particularly when it comes to identifying uneven or low-lying terrain. 2️⃣ The Isaac Perceptor System from NVIDIA: With the use of AI and many cameras, NVIDIA's Isaac Perceptor system is transforming robot vision. This makes it possible for robots to see in three dimensions, improving their comprehension of their surroundings. 3️⃣ Pioneers in the Industry: Additionally, businesses like Inovance Technology Europe and Boston Dynamics are expanding the capabilities of robot perception. Their sophisticated depth sensors and cameras help to enhance robot vision and intelligence. 4️⃣ Smart Cooperation: As robots' intelligence and vision grow, we are laying the groundwork for safer robot-human cooperation. Future developments might lead to more sophisticated and adaptable robots thanks to this breakthrough. Let's commemorate the incredible advancements in robot vision technology and the possible effects they may have on a range of sectors. Watch this space for further AI and tech upgrades! 🚀 #TechInnovation #AIProgress #RoboticsRevolution #FutureTech #TechProgress #RobotVision

[FLASH IMAGING] How to map instantaneously depth and speed at the same time? 👀 Commercial depth mapping applications (for Face ID type 🤳 or pedestrian detection on roads) today use the so-called Time of Flight technique based on the illumination of the scene by a laser. But these imagers do not allow the instantaneous speed measurement of moving objects at video rate. And are also not naturally immune to strong light conditions.  ➕ Advantages of our solution :    ·      No scanning   ·      Distance and speed measurement at video rate, with high resolution ·      Immunity to light conditions ·      Eye safetylow power consumption A game changer for : ·      autonomous vehicles 🚗 robotics 🤖 face recognition 👧 ·      AR/VR 🥽 Machine vision ⚙ Smart autofocus  ✊ Join us & be actor of this technological #breakthrough!  📍 Meet our experts in our booth #5129 at SPIE Photonics West, Jan. 30-Feb.1 - The Moscone Center - San Francisco Pierre Castelein | Vygintas Jankus | Vincent Destefanis | Eleonore Hardy | Laurent Fulbert | Estelle Fege | Aurélie Tapini | SPIE, the international society for optics and photonics #optics #photonics #3D #imaging #technology 

Take a look at the features and benefits of CenoBots L50 20” autoscrubber. Your wallet will be happy with our value pricing and ROI. Let’s chat today. #CenoBots #autoscrubber #schoolcleaning #robots #autonomouscleaning

How Robots See the World?! Robots use sensors to "see," with lidar being a key one. Lidar creates a 3D map using lasers but can miss low obstacles or uneven ground. NVIDIA's Isaac Perceptor system uses cameras and AI for a complete view, giving robots 3D vision to measure distance and understand objects. Other companies like Boston Dynamics and Inovance are also improving robot vision with advanced cameras and depth sensors. Smarter robots with better vision mean safer human-robot collaboration and more intelligent, versatile robots in the future. Follow me for AI updates. #ai #tech #robotics #innovation #gtc2024

Kicking off the year with a bit of SLAM! Simultaneous Localisation and Mapping, to be precise. 😁 This is similar to the work I do with cameras but with a special LiDAR scanner instead! The end result is a point cloud - a 3D model of whatever structure, equipment, or location that’s been scanned, made up of millions of points that can be used as the base model for a digital twin. It’s so easy to find a great workflow for something and just roll with it for months, years, or even decades. The argument often being ‘why change something that works?’. But so often, workflows don’t stand the test of time, especially in such an innovative world! I’m naturally a problem solver/lover of efficiency, so when the chance comes up to solve, enhance, or figure out how to make something better or more efficient, my hand always goes up to volunteer. Does anyone else absolutely LOVE making things more efficient?! 😍 So, I’ve been out with some of the team testing new processes and workflows to keep me and the team at the very top of our game. Remaining agile in the data capture and digital twin space is crucial and a big part of that is being flexible and adaptable to change, even when everything is already working well. And I think it’s safe to say, that translates across businesses that span all sectors and industries, but also just, the world in general. For where would the world be, without innovation? Share your thoughts below 👇 ZynQ 360 #innovation #digitaltwin #aberdeen #oilgas #heritage

Exciting news from Teledyne FLIR, a part of Teledyne Technologies Incorporated, as they introduce the Teledyne FLIR AVP, a state-of-the-art video processor powering their Prism AI and computational imaging at the edge. This groundbreaking technology integrates the latest Qualcomm QCS8550 mobile processor chip, ensuring top-notch AI performance in a compact, energy-efficient module. The AVP revolutionizes thermal and visible camera integration across various platforms such as unmanned aerial vehicles, robotics, gimbals, handheld devices, and security systems. It leverages Teledyne FLIR’s Prism AI and ISP software libraries, interfacing seamlessly with Boson and Neutrino thermal infrared imaging camera modules along with popular visible cameras. Prism AI, trained on a vast thermal image data lake, offers robust perception capabilities for applications ranging from automotive autonomy to security systems. On the other hand, Prism ISP delivers advanced image processing algorithms vital for image enhancement and analysis. According to Dan Walker, Vice President of Product Management at Teledyne FLIR, the AVP sets new benchmarks in power efficiency and AI workload processing, promising up to five times faster performance compared to competitors. This advancement not only enhances system capabilities but also simplifies development processes and reduces integration risks. To experience the AVP firsthand, visit Teledyne's booth #702 at the SPIE, the international society for optics and photonics, Defense Commercial Sensing Exhibition from April 23 to April 24, 2024. Stay updated with our blog for more insights into Teledyne FLIR’s cutting-edge technology and its impact across industries. #AvfoilNews #TeledyneFLIR #TeledyneTechnologies #PrismAI #AVP #InnovationAtTheEdge