NVIDIA Introduces NeuralVDB, Bringing AI and GPU Optimization to Award-Successful OpenVDB


NVIDIA at present introduced NeuralVDB, which brings the ability of AI to OpenVDB, the industry-standard library for simulating and rendering sparse volumetric knowledge, resembling water, fireplace, smoke and clouds.

Constructing on the previous decade’s growth of OpenVDB, the introduction at SIGGRAPH of NeuralVDB is a game-changer for professionals working in areas like scientific computing and visualization, medical imaging, rocket science and visible results. By decreasing reminiscence footprint by as much as 100x, it permits creators, builders and researchers to work together with extraordinarily giant and complicated datasets in actual time.

Over the previous decade, OpenVDB has earned Academy Awards as a core know-how used all through the visual-effects {industry}. It has since grown past leisure to industrial and scientific use circumstances the place sparse volumetric knowledge is prevalent, resembling industrial design and robotics.

Final 12 months, NVIDIA launched NanoVDB, which added GPU assist to OpenVDB. This delivered an order-of-magnitude speedup, enabling sooner efficiency and simpler growth — and opening the door to real-time simulation and rendering.

NeuralVDB builds on the GPU acceleration of NanoVDB by including machine studying to introduce compact neural representations that dramatically cut back its reminiscence footprint. This permits 3D knowledge to be represented at even greater decision and at a a lot bigger scale than OpenVDB. The result’s that customers can simply deal with huge volumetric datasets on units like particular person workstations and even laptops.

NeuralVDB provides a big effectivity enchancment over OpenVDB by compressing a quantity’s reminiscence footprint as much as 100x in comparison with NanoVDB. This permits customers to transmit and share giant, advanced volumetric datasets way more effectively.

To speed up coaching as much as 2x, NeuralVDB permits the weights of a body for use for the following one. NeuralVDB additionally permits customers to realize temporal coherency, or clean encoding, through the use of the community outcomes from the earlier body.

Hitting this trifecta of dramatically decreasing reminiscence necessities, accelerating coaching and enabling temporal coherency permits NeuralVDB to unlock new potentialities for scientific and industrial use circumstances, together with huge, advanced quantity datasets for AI-enabled medical imaging, large-scale digital twin simulations and extra.

Be taught extra about NeuralVDB.

Watch the NVIDIA particular deal with at SIGGRAPH on demand, and be a part of NVIDIA on the convention via Thursday, Aug. 11, to see extra of the most recent know-how breakthroughs in graphics, AI and digital worlds.



Source_link

Leave a Reply

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