Skip to content

POSTECH-IMLAB/IR_AnimalDetector

Repository files navigation

PIMNet_AnimalDetector

Created by EunSeop and Daijin Kim at POSTECH IM Lab

Overview

We propose a new IR based animal detection method for ADAS based on YOLOv2 that extracts features through base network and a hourglass method robust to small object detection. To improve the performance of Recall, we use anchor boxes to detect object.

How to use

You can easily train/test your IR images (or videos) using shell scripts (test_IR_animals.sh, train_IR_animals).

For training, you should download the pre-trained weight file from https://pjreddie.com/media/files/yolov2.weights

Acknowledgements

This research was supported by the MSIT (Ministry of Science, ICT), Korea, under the SW Starlab support program (IITP-2017-0-00897) supervised by the IITP (Institute for Information & communications Technology Promotion) and also supported by the MSIT, Korea, under the ICT Consilience Creative program (IITP-2017-R0346-16-1007) supervised by the IITP.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages