Senior Technical Director, Vision Systems Laboratory, Center for Vision Technologies
Michael Piacentino is technical director of the Vision Systems Laboratory in SRI International’s Center for Vision Technologies. He manages a team of more than 25 vision hardware and software architects in the development of state-of-the-art vision systems, including networking and embedded platforms.
Specializing in the development of real-time hardware and software solutions, Piacentino has been SRI’s embedded vision and system lead on numerous Defense Advanced Research Projects Agency (DARPA), Army, Navy, and other government and commercial programs. His professional competence includes hardware system and chip design for video processing algorithms, with emphasis on custom application-specific integrated circuit (ASIC) and electronics systems, video processing design, and algorithm implementation.
Previously, Piacentino was lead digital designer and system architect at Hughes Missile Systems. His responsibilities included developing and leading teams in the design of critical missile visible and RF guidance systems.
Piacentino has given more than 30 conference presentations, authored or co-authored more than 20 peer-reviewed technical papers, and holds five U.S. patents relating to computer vision hardware. He has also achieved numerous technical achievement awards at SRI, including the design and innovation of several world-leading vision chips and processors.
We present a scalable in-pixel processing architecture that can reduce the data throughput by 10X and consume less than 30 mW per megapixel at the imager frontend.
To enable learning on edge devices with fast convergence and low memory, we present a novel backpropagation-free optimization algorithm dubbed Target Projection Stochastic Gradient Descent (tpSGD).
In this paper we present Hyper-Dimensional Reconfigurable Analytics at the Tactical Edge using low-SWaP embedded hardware that can perform real-time reconfiguration at the edge leveraging non-MAC deep neural nets (DNN)…
We evaluate our proposed algorithms on three different scenarios comprising tasks from the Starcraft 2 and Minigrid domains.
We review HyDRATE, a low-SWaP reconfigurable neural network architecture developed under the DARPA AIE HyDDENN (Hyper-Dimensional Data Enabled Neural Network) program.
Automated Image Analysis and Classification Tool Based on Computer Vision Deep Learning Technologies
We present a rapid underwater video and automated image analysis tool using computer vision deep learning technologies.