LoCS-Net: Localizing convolutional spiking neural network for fast visual place recognition

Ugur Akcal, Ivan Georgiev Raikov, Ekaterina Dmitrievna Gribkova, Anwesa Choudhuri, Seung Hyun Kim, Mattia Gazzola, Rhanor Gillette, Ivan Soltesz, Girish Chowdhary · 2025

Read LoCS-Net: Localizing convolutional spiking neural network for fast visual place recognition by Ugur Akcal, Ivan Georgiev Raikov, Ekaterina Dmitrievna Gribkova, Anwesa Choudhuri, Seung Hyun Kim, Mattia Gazzola, Rhanor Gillette, Ivan Soltesz, Girish Chowdhary (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11813887.txt.

Description

Ugur Akcal and co-authors built a spiking neural network that recognizes places from images in a paper published in two thousand twenty-four. Robots use visual place recognition to tell where they are, but standard networks can be slow and power-hungry. Spiking networks, which fire like brain cells, promise big energy savings.

Their model, LoCS-Net, trains with smooth rate-based neurons, then swaps in spiking neurons for real use. It beat prior spiking models on the Nordland and Oxford RobotCar datasets, with large gains in precision. Tests on a neuromorphic chip from Intel kept most of that edge while using far less energy. The authors see a step toward practical robot navigation.

Catalog record

CREATORUgur Akcal, Ivan Georgiev Raikov, Ekaterina Dmitrievna Gribkova, Anwesa Choudhuri, Seung Hyun Kim, Mattia Gazzola, Rhanor Gillette, Ivan Soltesz, Girish Chowdhary
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE48 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC11813887

Browse the full library · About the Tower of Babel