Detecting and Tracking of Multiple Mice Using Part Proposal Networks

Abstract

The study of mouse social behaviours has been increasingly undertaken in neuroscience research. However, automated quantification of mouse behaviours from the videos of interacting mice is still a challenging problem, where object tracking plays a key role in locating mice in their living spaces. Artificial markers are often applied for multiple mice tracking, which are intrusive and consequently interfere with the movements of mice in a dynamic environment. In this paper, we propose a novel method to continuously track several mice and individual parts without requiring any specific tagging. Firstly, we propose an efficient and robust deep learning based mouse part detection scheme to generate part candidates. Subsequently, we propose a novel Bayesianinference Integer Linear Programming Model that jointly assigns the part candidates to individual targets with necessary geometric constraints whilst establishing pair-wise association between the detected parts. There is no publicly available dataset in the research community that provides a quantitative test-bed for the part detection and tracking of multiple mice, and we here introduce a new challenging Multi-Mice PartsTrack dataset that is made of complex behaviours. Finally, we evaluate our proposed approach against several baselines on our new datasets, where the results show that our method outperforms the other state-of-the-art approaches in terms of accuracy. We also demonstrate the generalization ability of the proposed approach on tracking zebra and locust.

Extendable Multiple Nodes Recurrent Tracking Framework With RTU++

“…Afshin et al [25] formulate the data association during tracking as maximum multiple cliques problem. Jiang et al [26] design a linear programming algorithm to finds the best trajectories. Kim et al [6] reconstruct the multi-object tracking by maximum weight independent set and introduce an online updated appearance model in their trackers.…”

Challenges of a small world analysis for the continuous monitoring of behavior in mice

“…Deep learning has been successfully applied to this form of pose estimation in humans and other animals . This approach uses several pipelines to perform multiple-mice tracking for the study of social interactions (Jiang et al, 2019b). The authors decompose the problem of tracking One limitation of automated supervised approaches for behavioral classification and the study of social interactions is that all approaches require labeled examples to train the models.…”

Automation of Computer Vision Applications for Real-time Combat Sports Video Analysis

“…Kalake et al (2022) propose a paradigm aimed at eliminating object tracking difficulties by enhancing the detection quality rate through the combination of a convolutional neural network (CNN) and a histogram of oriented gradient (HOG) descriptor. Jiang et al (2022) propose a novel method to continuously track several mice and individual parts without requiring any specific tagging. 2021) present an end-to-end trainable approach for joint object detection and tracking that is capable of object permanence and approximate object localization in the presence of full occlusions.…”