24-GOPS 4.5-mm(2) Digital Cellular Neural Network for Rapid Visual Attention in an Object-Recognition SoC

Cited 29 time in webofscience Cited 0 time in scopus
  • Hit : 496
  • Download : 24
This paper presents the Visual Attention Engine (VAE), which is a digital cellular neural network (CNN) that executes the VA algorithm to speed up object-recognition. The proposed time-multiplexed processing element (TMPE) CNN topology achieves high performance and small area by integrating 4800 (80 x 60) cells and 120 PEs. Pipelined operation of the PEs and single-cycle global shift capability of the cells result in a high PE utilization ratio of 93%. The cells are implemented by 6T static random access memory-based register files and dynamic shift registers to enable a small area of 4.5 mm(2). The bus connections between PEs and cells are optimized to minimize power consumption. The VAE is integrated within an object-recognition system-on-chip (SoC) fabricated in the 0.13-mu m complementary metal-oxide-semiconductor process. It achieves 24 GOPS peak performance and 22 GOPS sustained performance at 200 MHz enabling one CNN iteration on an 80 x 60 pixel image to be completed in just 4.3 mu s. With VA enabled using the VAE, the workload of the object-recognition SoC is significantly reduced, resulting in 83% higher frame rate while consuming 45% less energy per frame without degradation of recognition accuracy.
Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Issue Date
2011-01
Language
English
Article Type
Article
Citation

IEEE TRANSACTIONS ON NEURAL NETWORKS, v.22, no.1, pp.64 - 73

ISSN
1045-9227
DOI
10.1109/TNN.2010.2085443
URI
http://hdl.handle.net/10203/98615
Appears in Collection
EE-Journal Papers(저널논문)
Files in This Item
This item is cited by other documents in WoS
⊙ Detail Information in WoSⓡ Click to see webofscience_button
⊙ Cited 29 items in WoS Click to see citing articles in records_button

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0