A Self-Supervised Method for Body Part Segmentation and Keypoint Detection of Rat Images

TL;DR

This paper introduces a self-supervised deep learning approach for rat body part segmentation and keypoint detection, reducing manual annotation efforts and effectively handling occlusions in laboratory animal behavior analysis.

Contribution

It presents a novel method that generates initial annotations automatically and trains a neural network, enabling accurate segmentation and keypoint detection without extensive manual labeling.

Findings

Effective segmentation and keypoint detection on rat images

Handles heavy occlusions in laboratory settings

Reduces need for manual annotation

Abstract

Recognition of individual components and keypoint detection supported by instance segmentation is crucial to analyze the behavior of agents on the scene. Such systems could be used for surveillance, self-driving cars, and also for medical research, where behavior analysis of laboratory animals is used to confirm the aftereffects of a given medicine. A method capable of solving the aforementioned tasks usually requires a large amount of high-quality hand-annotated data, which takes time and money to produce. In this paper, we propose a method that alleviates the need for manual labeling of laboratory rats. To do so, first, we generate initial annotations with a computer vision-based approach, then through extensive augmentation, we train a deep neural network on the generated data. The final system is capable of instance segmentation, keypoint detection, and body part segmentation even when the objects are heavily occluded.