Abaco Systems’ Post

Integrate multiple applications. Simultaneously integrate and access numerous applications, including sensor targeting, moving maps, mission computers and instrumentation graphics processing, with displays built utilizing open architectures and diverse I/O. #EuropeanRotors23 #InnovationThatMatters

Integrate multiple applications. Simultaneously integrate and access numerous applications, including sensor targeting, moving maps, mission computers and instrumentation graphics processing, with displays built utilizing open architectures and diverse I/O. #EuropeanRotors23 #InnovationThatMatters

Full video demonstrating of realtime rendering of Revit scene using Intel Arc Pro A40 GPUs for workstations using #GenAI! In this proof-of-concept, I'm using a simple wireframe preview in Revit, from Autodesk, as realtime input for stable diffusion image generation. All on an #IntelArc A40 GPU with great results. Architects and designers could leverage GenAI as shown to quickly iterate concept renders using a variety of styles and materials without additional cycles in modelling and material workflow. Watch here

More brain, less power That’s what you can expect from our latest SMARC-Modules based on the NXP Semiconductors i. MX 95 processor series: 🧠 Integrated NXP eIQ Neutron NPU for AI inference and machine learning at the edge  💡 Powerful NXP eIQ® development environment facilitates the creation of AI applications  🌐 Brand new Arm Mali 3D graphics for outstanding GPU performance  🌡️ Industrial temperature range from -40 °C to +85 °C Ideal for AI-accelerated low-power applications in industrial production, machine vision and robotics. Find out more about our new SMARC modules and their wide application range here: https://bit.ly/3XV6TYb #congatec #EmbeddedSystems #EdgeComputing #NXP #Innovation #imx95 #SMARC #SMARCmodule #Arm #armcortex #computeronmodules

The another new #SMARC computer-on-module product line from #congatec based on #Arm application processor from #NXP #imx95 with embedded #machinelearning accelerator aka NPU to assist #ai applications with #Arm. Its NPU performance has gained x2 from the previous one i.MX8M-Plus. Personally exciting to scale my experience at my current and a previous company on i.MX8M-Plus into the next generation, which includes building up demo of eIQ (#NXP #ai toolkit) as well as industrial network (#CAN and #TSN) that are also available on #imx95 thus #congatec #SMARC module aka conga-SMX95. #EmbeddedSystems #EdgeComputing #Innovation hashtag #armcortex hashtag #computeronmodules #industry4

Dive into the wild world of decentralized GPU rendering with our latest paper! 🦌 Explore how the Render Network harnesses the power of idle GPUs to revolutionize image processing. 🎨✨ Discover the magic of decentralized rendering and unleash your creative potential! #RenderNetwork #DecentralizedRendering #WildRide https://lnkd.in/gTk8nqrT

Transformer models for 3D Object Detection not only improve the detection accuracy and deliver more consistent detection results, but also help us to improve the inference performance in our perception stack. Our End-to-End Transformer detector, optimized with TensorRT, reaches 38 FPS on a Nvidia RTX 2080, while the PointPillars model including a Rotated-NMS stays 8 FPS behind. A special note of gratitude to Erik Deinzer for his excellent bachelor thesis under my supervision and the guidance of Fabian Thomsen. His significant contribution includes the implementation, training and optimization of a transformer-based 3D Object Detector in our machine learning framework. #computervision #objectdetection #detr #vit #transformer

NVIDIA GTC has been a blast so far – we’re very happy to be able to demonstrate our cutting-edge tech at the conference for the era of AI, computer graphics, data science, and more.    Thanks to Jensen Huang for stopping by Kinetic Vision’s booth and checking out #NavVisVLX, the tool used to help create a digital twin of their next-gen data center. We’re on-site for the final day, so make sure to swing by booth 531.    To read more about our partnership with NVIDIA & Kinetic Vision, check out the full press release here: https://lnkd.in/dwetJBam #BuildBetterReality #NVIDIAGTC #IndustrialMetaverse #NVIDIA #Omniverse #GTC2024 #DigitalTwin #GTC24 #NVIDIAOmniverse #OpenUSD 

the future of graphics processing is absolutely incredible

Finally found some time to peak into the new #particles in Maxon #C4D The performance is quite impressive, as long as you don´t make the system download data from GPU to CPU. Unfortunately using fields does utilize the CPU, therefore you can´t really avoid it in more complex setups at the moment. No idea, if this can be transferred on the GPU in future releases, but would make things much faster. What I really like about the system is, how well it plays with the existing toolset in the app. Not only the simulation system, but also Mograph. You want geometry on particles? Use a cloner. You want to generate splines? Use a tracer. No redundancy here, just integration in the known toolset. I am really looking forward to future releases and hope the development will continue. There is still a lot missing (like a FLIP solver or advanced particle2particle behaviour in general), but the first release is promising :) #3D #motiongraphics #motiondesign #simulation #cinema4D