DeeperCut: A Deeper, Stronger, and Faster Multi-Person Pose Estimation Model

Insafutdinov, Eldar; Pishchulin, Leonid; Andres, Bjoern; Andriluka, Mykhaylo; Schiele, Bernt

doi:10.48550/arxiv.1605.03170

Cited by 15 publications

(44 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We use DeeperCut [6] to estimate the location of human body parts in each color frame of the video. Figure 1 shows a video frame where we have superimposed the body part locations estimated by DeeperCut.…”

Section: Our Methodsmentioning

confidence: 99%

“…None of the 15 subjects appearing in the test set is used to train any part of our models. The only module that uses training is DeeperCut, and we use the pretrained model that has been made available by the authors of [6].…”

Section: Methodsmentioning

confidence: 99%

“…Several deep-learning methods have been proposed in recent years for video analysis and activity recognition [1,4,2], offering significantly improved accuracy compared to previous approaches [7,10]. Deep learning methods have also been used in supervised or unsupervised manner in different tasks in computer vision [6,5], oftentimes producing state-of-the-art results.…”

Section: Related Workmentioning

confidence: 99%

“…In CogniLearn, a Microsoft Kinect V2 camera is used for recording human motion. Then, we use the DeeperCut method [6] to perform body pose estimation in each frame. Finally, using the body pose estimates from DeeperCut we use a classification module that determines whether the subject touched his or her head, shoulders, knees, or toes.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Improving the Accuracy of the CogniLearn System for Cognitive Behavior Assessment

Ghaderi

Gattupalli

Ebert

et al. 2017

Proceedings of the 10th International Conference on PErvasive Technologies Related to Assistive Environments

View full text Add to dashboard Cite

HTKS [9] is a game-like cognitive assessment method, designed for children between four and eight years of age. During the HTKS assessment, a child responds to a sequence of requests, such as "touch your head" or "touch your toes". The cognitive challenge stems from the fact that the children are instructed to interpret these requests not literally, but by touching a different body part than the one stated. In prior work, we have developed the CogniLearn system, that captures data from subjects performing the HTKS game, and analyzes the motion of the subjects. In this paper we propose some specific improvements that make the motion analysis module more accurate. As a result of these improvements, the accuracy in recognizing cases where subjects touch their toes has gone from 76.46% in our previous work to 97.19% in this paper.

show abstract

Section: Our Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Improving the Accuracy of the CogniLearn System for Cognitive Behavior Assessment

Ghaderi

Gattupalli

Ebert

et al. 2017

Proceedings of the 10th International Conference on PErvasive Technologies Related to Assistive Environments

View full text Add to dashboard Cite

show abstract

“…They decreased the depth of the network and increased the size of a Residual Module by adding more features and convolutional layers. Currently, ResNets are state-of-the-art ConvNet models and they have been shown to perform remarkable well both in image recognition [19][20][21] and human pose estimation tasks [15,22].…”

Section: Introductionmentioning

confidence: 99%

Real-time Human Pose Estimation with Convolutional Neural Networks

Linna

Kannala

Rahtu

2018

Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Application

View full text Add to dashboard Cite

In this paper, we present a method for real-time multi-person human pose estimation from video by utilizing convolutional neural networks. Our method is aimed for use case specific applications, where good accuracy is essential and variation of the background and poses is limited. This enables us to use a generic network architecture, which is both accurate and fast. We divide the problem into two phases: (1) pre-training and (2) finetuning. In pre-training, the network is learned with highly diverse input data from publicly available datasets, while in finetuning we train with application specific data, which we record with Kinect. Our method differs from most of the state-of-the-art methods in that we consider the whole system, including person detector, pose estimator and an automatic way to record application specific training material for finetuning. Our method is considerably faster than many of the state-of-the-art methods. Our method can be thought of as a replacement for Kinect, and it can be used for higher level tasks, such as gesture control, games, person tracking, action recognition and action tracking. We achieved accuracy of 96.8% (PCK@0.2) with application specific data.

show abstract