• Corpus ID: 37545044

Automated Mouse Behavior Recognition using VGG Features and LSTM Networks

  title={Automated Mouse Behavior Recognition using VGG Features and LSTM Networks},
  author={Gregory Kramida and Yiannis Aloimonos and Cornelia Ferm{\"u}ller and Nikolas A. Francis},
We present a mouse behavior classification method using a recurrent neural network with the long short-term memory (LSTM) model. The experimental hardware used to collect the data is a custom mouse cage with four stereo-camera pairs in each wall. Using as input the different videos, our computational method employs a so-called end-to-end learning approach: visual features from pre-trained convolutional neural networks are extracted from each image frame, and used to train a customized LSTM… 

Figures from this paper

Recurrent 3D Convolutional Network for Rodent Behavior Recognition
  • Van Anh Le, K. Murari
  • Computer Science
    ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2019
A deep architecture which is a combination of recurrent network and 3D convolutional network to learn long and short-term video representations is proposed, demonstrating that the framework can obtain accuracy on par with human assessment.
Context-Aware Mouse Behavior Recognition Using Hidden Markov Models
This paper develops and implements a novel hidden Markov model (HMM) algorithm to describe the temporal characteristics of mouse behaviors, and proposes a hybrid deep learning architecture, where the first unsupervised layer relies on an advanced spatial-temporal segment Fisher vector encoding both visual and contextual features.
Semi-supervised sequence modeling for improved behavioral segmentation
It is shown that a large number of unlabeled frames can improve supervised segmentation in the regime of sparse hand labels and also show that a small number of hand labeled frames can increase the precision of unsupervised segmentation.
Automated Recognition of Rodent Social Behavior
It is demonstrated that placing the human in the annotation loop reduces the annotation time substantially compared to traditional, sequential labeling, and this interactive annotation approach enables neuroscientists to analyze behavioral data quicker than before and to study previous data in new light with limited manual work.


Very Deep Convolutional Networks for Large-Scale Image Recognition
This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.
Automated home-cage behavioural phenotyping of mice.
A trainable computer vision system enabling the automated analysis of complex mouse behaviours that performs on par with human scoring, as measured from ground-truth manual annotations of thousands of clips of freely behaving mice.
Automated measurement of mouse social behaviors using depth sensing, video tracking, and machine learning
An integrated hardware and software system that combines video tracking, depth-sensing technology, machine vision, and machine learning to automatically detect and score innate social behaviors, including aggression, mating, and social investigation, between mice in a home-cage environment is described.
Automatic Visual Tracking and Social Behaviour Analysis with Multiple Mice
The validity and effectiveness of this new machine learning system in the detection of social and non-social behaviours in multiple (>2) interacting mice, and its versatility to deal with different experimental settings and scenarios are demonstrated.
Dropout: a simple way to prevent neural networks from overfitting
It is shown that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets.
Caffe: Convolutional Architecture for Fast Feature Embedding
Caffe provides multimedia scientists and practitioners with a clean and modifiable framework for state-of-the-art deep learning algorithms and a collection of reference models for training and deploying general-purpose convolutional neural networks and other deep models efficiently on commodity architectures.
ADADELTA: An Adaptive Learning Rate Method
We present a novel per-dimension learning rate method for gradient descent called ADADELTA. The method dynamically adapts over time using only first order information and has minimal computational
Long Short-Term Memory
A novel, efficient, gradient based method called long short-term memory (LSTM) is introduced, which can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units.
Independent multimodal background subtraction
A novel real-time algorithm for creating a robust and multimodal background model is presented based on an on-line clustering algorithm to create the model and on a novel conditional update mechanism that allows for obtaining an accurate foreground mask.
Theano: A CPU and GPU Math Compiler in Python
This paper illustrates how to use Theano, outlines the scope of the compiler, provides benchmarks on both CPU and GPU processors, and explains its overall design.