DeepPoseKit, a software toolkit for fast and robust animal pose estimation using deep learning

Graving, Jacob M.; Chae, Daniel H.; Naik, Hemal; Li, Liang; Koger, Benjamin; Costelloe, Blair R.; Couzin, Iain D.

doi:10.7554/elife.47994

articleeLifeSep 30, 2019GOLD OA

DeepPoseKit, a software toolkit for fast and robust animal pose estimation using deep learning

JMJacob M. Graving DHDaniel H. Chae HNHemal Naik LLLiang Li BKBenjamin Koger

University of Konstanz · Max Planck Institute of Animal Behavior · +2 more institutions

PubMed

Indexed incrossrefdoajpubmed

Abstract

Quantitative behavioral measurements are important for answering questions across scientific disciplines—from neuroscience to ecology. State-of-the-art deep-learning methods offer major advances in data quality and detail by allowing researchers to automatically estimate locations of an animal’s body parts directly from images or videos. However, currently available animal pose estimation methods have limitations in speed and robustness. Here, we introduce a new easy-to-use software toolkit, DeepPoseKit, that addresses these problems using an efficient multi-scale deep-learning model, called Stacked DenseNet, and a fast GPU-based peak-detection algorithm for estimating keypoint locations with subpixel…

Citation impact

562

total citations

FWCI: 72.30
Percentile: 100%
References: 145

Citations per year

Authors

7

Topics & keywords

Topics

Keywords

Subpixel rendering
Computer science
Robustness (evolution)
Artificial intelligence
Deep learning
Machine learning
Software
Pose

No related works found for this paper.