Pose Estimation
pose estimation
Ann Kennedy
We investigate principles of computation in meso-scale biological neural networks, and the role of these networks in shaping animal behavior. We work in collaboration with experimental neuroscientists recording neural activity in freely moving animals engaged in complex behaviors, to investigate how animals' environments, actions, and internal states are represented across multiple brain areas. Our work is especially inspired by the interaction between subcortical neural populations organized into heavily recurrent neural circuits, including basal ganglia and nuclei of the hypothalamus. Project in the lab include 1) developing novel supervised, semi-supervised, and unsupervised approaches to studying the structure of animal behavior, 2) using behavior as a common basis with which to model the interactions between multiple brain areas, and 3) studying computation and dynamics in networks of heterogenous neurons communicating with multiple neuromodulators and neuropeptides. The lab will also soon begin collecting behavioral data from freely interacting mice in a variety of model lines and animal conditions, to better chart the space of interactions between animal state and behavior expression. Come join us!
Monitoring gait outcomes in rehabilitation with human pose estimation and wearable sensors
DeepLabStream
DeepLabStream is a python based multi-purpose tool that enables the realtime tracking and manipulation of animals during ongoing experiments. Our toolbox was orginally adapted from the previously published DeepLabCut (Mathis et al., 2018) and expanded on its core capabilities, but is now able to utilize a variety of different network architectures for online pose estimation (SLEAP, DLC-Live, DeepPosekit's StackedDenseNet, StackedHourGlass and LEAP). Our aim is to provide an open-source tool that allows researchers to design custom experiments based on real-time behavior-dependent feedback. My personal ideal goal would be a swiss-army knife like solution where we could integrate the many brilliant python interfaces. We are constantly upgrading DLStream with new features and integrate other open-source solutions.
SemiMultiPose: A Semi-supervised Multi-animal Pose Estimation Framework
COSYNE 2022
SemiMultiPose: A Semi-supervised Multi-animal Pose Estimation Framework
COSYNE 2022
Pose estimation made better, easier, and faster with video semi-supervised learning on the cloud
COSYNE 2023
Quantifying social behaviors in juvenile Shank3 mice using animal pose estimation tools
FENS Forum 2024
Automated Research Workflows for Pose Estimation
Neuromatch 5