Facial Recognition Overview


Facial recognition systems are increasingly standard in a wide range of settings: from smart airports to automated retail experiences and beyond.

Effective facial recognition systems must extract accurate results with minimum latency between data capture and positive identification and be able to handle multiple queries almost simultaneously.

However, current solutions require very large numbers of costly GPUs and CPUs to deliver performance at scale.

The market seeks a new solution that can meet performance expectations while keeping infrastructure and TCO (Total Cost of Ownership) low.



Introducing a new facial recognition search engine powered by GSI’s Gemini® Associative Processing Unit (APU) on the GSI APU board.

Gemini® delivers accurate search results while reducing search times in large databases from many minutes to fractions of a second,
all with very low power consumption.



Fast and Accurate

The GSI APU is a true compute-in-memory device that inherently excels in associative parallel search. GSI applications can utilize this capability to support fast, 100% accurate KNN searches in the thousands of parameters per chip, or to balance the workload with the host processor to provide billion scale ANN searches per chip.

The GSI APU expands like memory in a server, so additional resources just require more APU PCIe cards to be installed in the server without expensive, high-speed interfaces between them. This allows users to tune their usage for database size, number of characteristics, accuracy, and latency depending on their application budget.




Scalable and Low Power

GSI’s face recognition search engine can process 32-bit and 64-bit floating-point vectors with 128 features and above (no upper limit).

A scalable solution, large datasets and multiple queries can be processed using one or multiple APU boards, with no loss of performance and with very low power consumption—approximately 3.5 times lower power consumption when compared with CPU-only systems.

Additionally, the GSI solution supports zero-shot learning, whereby a new image category can be recognized by inserting the target vector into the search database.






Facial Recognition Products

1-8 Leda-E 2U Server Contact for Purchase
16 Leda-S 1U Server Contact for Other Configurations
Leda-E PCIe card  Contact for Purchase
Hosted Service Contact for Details



Facial Recognition Resources


Gemini APU: Enabling High Performance
Billion-Scale Similarity Search

Read More…


Fast Visual Search in Hamming

Read More…


Seven Tips for Visual Search at

Read More…


The Visual Similarity Search

Read More…


Visual Search: The Future of

Read More…


High-Performance, Billion-Scale
Similarity Search

Read More…


Similarity Search: Finding a Needle in a

Read More…


Leveraging Word2vec For More Than

Read More…


Wayfair’s Visual Search and the Latest
in Similarity Search Research

Read More…



©2024 GSI Technology, Inc. All Rights Reserved