Learning effective human pose estimation from inaccurate annotation

By Sam Johnson, Mark Everingham. 2011

476 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Towards Accurate Multi-person Pose Estimation in the Wild (2017, neutral): “D human pose inference [3, 12, 37, 45, 11, 27, 32, 38, 18].”
Convolutional Pose Machines (2016, methodology): “We evaluate our method on the Extended Leeds Sports Dataset [15] that consists of 11000 images for training and 1000 images for testing.”
Extreme Clicking for Efficient Object Annotation (2017, background): “A qualification test is a good mechanism for enhancing the quality of crowd-sourcing data and for filtering out bad annotators and spammers [1, 19, 33, 37].”
DeepPose: Human Pose Estimation via Deep Neural Networks (2013, methodology): “Mixture models on the full model scale, by having mixture of PSs, have been studied by Johnson and Everingham [13].”
Stacked Hourglass Networks for Human Pose Estimation (2016, background): “Early work tackles such diﬃculties using robust image features and sophisticated structured prediction [1,2,3,4,5,6,7,8,9]: the former is used to produce local interpretations, whereas the latter is used to infer a globally consistent pose.”
Recurrent Human Pose Estimation (2016, methodology): “The evaluation is based on the MPII Human Pose [1] and LSP [21] datasets.”
CliqueCNN: Deep Unsupervised Exemplar Learning (2016, methodology): “8, Leeds Sports in Fig.”
Toward fast and accurate human pose estimation via soft-gated skip connections (2020, methodology): “The dataset was later expanded in [18] with 10,000 more images for training (LSP-Extended).”
Learning to Estimate 3D Human Pose and Shape from a Single Color Image (2018, methodology): “The Human2D network is trained on MPII [4], LSP [19] and LSP-extended [20] data, using the silhouettes from Lassner et al.”
Dynamic Kernel Distillation for Efficient Pose Estimation in Videos (2019, unknown): “Thereby, we follow the conventions of still-image based pose estimation [17, 7, 31, 28], and also adopt another metric that takes torso size as reference distance.”
Deep Learning-based Human Pose Estimation: A Survey (2020, methodology): “Name Year Single-Person /Multi-Person Joints Number of images Evaluation protocol Train Val Test LSP [91] 2010 Single 14 1k 1k PCP LSP-extended [92] 2011 Single 14 10k PCP FLIC [214] 2013 Single 10 5k 1k PCP&PCK FLIC-full [214] 2013 Single 10 20k PCP&PCK FLIC-plus [233] 2014 Single 10 17k PCP&PCK”
PoseTrack: A Benchmark for Human Pose Estimation and Tracking (2017, background): “The most popular benchmarks to date for evaluation of single person pose estimation are “LSP” [25] (+ “LSP Extended” [26]), “MPII Human Pose (Single Person)” [1] and MS COCO Keypoints Challenge [28].”
MODEC: Multimodal Decomposable Models for Human Pose Estimation (2013, methodology): “Methods such as [11, 13] calibrate the models post-hoc using cross-validation data.”