Fast and Efficient Model for Real-Time Tiger Detection In The Wild

TL;DR

This paper introduces TigerNet, a lightweight FPN-based model for real-time Amur Tiger detection in the wild, capable of running on edge devices with high accuracy, and employs semi-supervised learning to enhance performance.

Contribution

The paper presents TigerNet, a simple, efficient detection network with semi-supervised training, optimized for edge devices and superior performance in tiger detection tasks.

Findings

TigerNet requires only 0.071 GFLOPs per image.

The model has 600k parameters, enabling real-time detection.

It outperforms other methods on the ATRW-ICCV 2019 tiger detection challenge.

Abstract

The highest accuracy object detectors to date are based either on a two-stage approach such as Fast R-CNN or one-stage detectors such as Retina-Net or SSD with deep and complex backbones. In this paper we present TigerNet - simple yet efficient FPN based network architecture for Amur Tiger Detection in the wild. The model has 600k parameters, requires 0.071 GFLOPs per image and can run on the edge devices (smart cameras) in near real time. In addition, we introduce a two-stage semi-supervised learning via pseudo-labelling learning approach to distill the knowledge from the larger networks. For ATRW-ICCV 2019 tiger detection sub-challenge, based on public leaderboard score, our approach shows superior performance in comparison to other methods.