Low-Power Image Recognition Challenge (LPIRC) 2017

 

LPIRC 2017 Competition in Honolulu, Hawaii, USA on 21 July 2017

The third IEEE International Low-Power Image Recognition Challenge (LPIRC) concluded on July 21 in Honolulu, co-located with International Conference on Computer Vision and Pattern Recognition (CVPR). The champion is Seoul National University (South Korea). The second prize is KPST and ETRI (South Korea). The third prize goes to Watrix (China).

Two additional teams received special prizes for participation. The first team is Florida International University and University of Maimi; the second team is University of Pittsburgh and LGE.

The 2017 Champion's score is

mAP (mean average precision)Energy (WH)Recognized ImagesScore (mAP / energy)
0.248382.0818819,7860.1193

This score is 2.7 time better than the 2016 LPIRC Champion's.

The 2017 LPIRC is sponsored by IEEE Rebooting Computing and IEEE GreenICT.

LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo LPIRC photo

Hawaii

For inquiries, please contact rcinfo@ieee.org.

Access the competition software here: https://github.com/ieeelpirc

Main Goal

Detect all relevant objects in as many images as possible of a common test set from the ImageNet object detection data set within 10 minutes.

 

Scoring System

Final score is the image processing score divided by the energy consumption.

Scoring System

 

Motivation

Many mobile systems (smartphones, electronic glass, autonomous robots) can capture images. These systems use batteries and energy conservation is essential. This challenge aims to discover the best technology in both image recognition and energy conservation. Winners will be evaluated based on both high recognition accuracy and low power usage.

 

Image Recognition

Image recognition involves many tasks. This challenge focuses on object detection, a basic routine in many recognition approaches. The following two examples illustrate the task. In the first example, there are two objects: a bird and a frog. In the second example, there are several objects: cars, persons, motorcycle, and a helmet. The training and validation data for LPIRC comes from the ImageNet Large Scale Visual Recognition Challenge detection competition. The test data will be specific to LPIRC.

Picture

Related Publications

 

Sponsors

IEEE Rebooting Computing IEEE Green ICT

 

Previous Competitions