Open Source Initiative To Speed Up Cloud Applications For


With accelerated image pre- and post-processing methods and tools, CV-CUDA can handle 10x as many photos for the same price.

NVIDIA introduced CV-CUDA, an open source toolkit for building accelerated end-to-end computer vision and image processing pipelines with the goal of supporting image processing that is both faster and more effective at scale. Videos make up the bulk of internet traffic. This film will increasingly feature computer graphics and AI special effects.

Adding to this complexity are the rising cloud computing costs and inefficiencies in the computer vision and AI-based image processing pipelines of fast-growing social media and video-sharing platforms. AI special effects like relight, rest, background blur and super resolution are accelerated with CV-CUDA. NVIDIA GPUs are already accelerating the inference phase of AI computer vision pipelines. However, pre- and post-processing using conventional computer vision technologies consumes time and resources.

Also Read :  Meet AiDice: An Algorithm for Large-Scale Anomaly Detection with AIOps in Azure Cloud

CV-CUDA provides developers with 50+ high-performance computer vision algorithms, a development framework that simplifies designing unique kernels, and zero-copy interfaces to remove bottlenecks in the AI ​​pipeline. Higher throughput and lower cloud computing costs are the end result. On a single GPU, CV-CUDA can handle 10x more streams.

All of this enables developers to work significantly faster on projects including video content development, 3D environments, image-based recommender systems, image recognition and video conferencing. Video content creation platforms must filter, enhance, and regulate millions of video streams every day while ensuring mobile users have the best possible experience with their apps on any phone.

  • It is hoped that CV-CUDA will allow Jobs to help create or extend 3D worlds and their components for those developing 3D worlds or Metaverse apps.
  • Mobile users can now leverage sophisticated and responsive image recognition applications thanks to CV-enabled CUDAs to significantly accelerate hyperscale image analysis and recognition pipelines.
  • In addition, CV-CUDA can handle complex augmented reality based functions in video conferences. These capabilities can entail complicated AI processing chains that require multiple pre- and post-processing processes.
Also Read :  Hackathon sponsors provide opportunities for Kennesaw State students

CV-CUDA accelerates pre- and post-processing pipelines with manually optimized CUDA kernels. It also connects seamlessly with popular deep learning frameworks like PyTorch and C/C++. In NVIDIA Omniverse, a virtual world collaboration and simulation platform for 3D workflows, CV-CUDA will be one of the key technologies helping to accelerate AI operations. In December, the code will be made available to developers ahead of the beta release in March.

Also Read :  HFCL launches World’s First Open source Wi-Fi 7 Access Points at India Mobile Congress





Source link