 |
ML Newsnew | past | best | rss | submit | about |
| SNN Based Region Proposal Networks for Neuromorphic Vision Sensors (arxiv.org) | |
| 2 points by godelmachine on Feb 27, 2019 | hide | past | pdf | discuss on HN | |
|
In plain words: A three-layer network of brain-like firing neurons spots moving objects in an event camera's stream of brightness changes, which records only what moves. On traffic footage it caught about 50% more real objects than the usual clustering approach, with similar accuracy and computing cost. Abstract · Spiking Neural Network based Region Proposal Networks for Neuromorphic Vision SensorsThis paper presents a three layer spiking neural network based region proposal network operating on data generated by neuromorphic vision sensors. The proposed architecture consists of refractory, convolution and clustering layers designed with bio-realistic leaky integrate and fire (LIF) neurons and synapses. The proposed algorithm is tested on traffic scene recordings from a DAVIS sensor setup. The performance of the region proposal network has been compared with event based mean shift algorithm and is found to be far superior (~50% better) in recall for similar precision (~85%). Computational and memory complexity of the proposed method are also shown to be similar to that of event based mean shift | |
| add comment on HN |