An 86 mW 98GOPS ANN-Searching Processor for Full-HD 30 fps Video Object Recognition With Zeroless Locality-Sensitive Hashing Gyeonghoon Kim, Jinwook Oh, Seungjin Lee, and Hoi-Jun Y oo

Abstract

Approximate nearest neighbor (ANN) searching is an essential task in object recognition. The ANN-searching stage, however, is the main bottleneck in the object recognition process due to increasing database siz e and massive dimensions of key- point descriptors. In this paper, a high throughput ANN-searching processor is proposed for high-resolution (full-HD) and real-time (30 fps) video object recognition. The proposed ANN-searching processor adopts an interfram e cache architecture as a hard- ware-oriented approach and a zerol ess locality-sensitive-hashing (zeroless-LSH) algorithm as a so ftware-oriented approach to re- duce the external memory bandwidth required in nearest neighbor searching. A four-way set associative on-chip cache has a dedi- cated architecture to exploit data correlation at the frame-level. Zeroless-LSH minimizes data tra nsactions from external memory at the vector-level. The proposed ANN-searching processor is fabricated as part of an object recognition SoC using a 0.13 6 metal CMOS technology. It achieves 62 720 vectors/s throughput and 1140 GOPS/W power ef ficiency, which are 1.45 and 1.37 times higher than the state-of-the-art, respectively, enabling real-time object recognition for full -HD 30 fps video streams.