scholarly journals Fast animal pose estimation using deep neural networks

2018 ◽  
Author(s):  
T.D. Pereira ◽  
D. E. Aldarondo ◽  
L. Willmore ◽  
M. Kislin ◽  
S. S.-H. Wang ◽  
...  

AbstractRecent work quantifying postural dynamics has attempted to define the repertoire of behaviors performed by an animal. However, a major drawback to these techniques has been their reliance on dimensionality reduction of images which destroys information about which parts of the body are used in each behavior. To address this issue, we introduce a deep learning-based method for pose estimation, LEAP (LEAP Estimates Animal Pose). LEAP automatically predicts the positions of animal body parts using a deep convolutional neural network with as little as 10 frames of labeled data for training. This framework consists of a graphical interface for interactive labeling of body parts and software for training the network and fast prediction on new data (1 hr to train, 185 Hz predictions). We validate LEAP using videos of freely behaving fruit flies (Drosophila melanogaster) and track 32 distinct points on the body to fully describe the pose of the head, body, wings, and legs with an error rate of <3% of the animal’s body length. We recapitulate a number of reported findings on insect gait dynamics and show LEAP’s applicability as the first step in unsupervised behavioral classification. Finally, we extend the method to more challenging imaging situations (pairs of flies moving on a mesh-like background) and movies from freely moving mice (Mus musculus) where we track the full conformation of the head, body, and limbs.

2019 ◽  
Vol 9 (3) ◽  
pp. 400 ◽  
Author(s):  
Zaiqiang Wu ◽  
Wei Jiang ◽  
Hao Luo ◽  
Lin Cheng

Statistical body shape models are widely used in 3D pose estimation due to their low-dimensional parameters representation. However, it is difficult to avoid self-intersection between body parts accurately. Motivated by this fact, we proposed a novel self-intersection penalty term for statistical body shape models applied in 3D pose estimation. To avoid the trouble of computing self-intersection for complex surfaces like the body meshes, the gradient of our proposed self-intersection penalty term is manually derived from the perspective of geometry. First, the self-intersection penalty term is defined as the volume of the self-intersection region. To calculate the partial derivatives with respect to the coordinates of the vertices, we employed detection rays to divide vertices of statistical body shape models into different groups depending on whether the vertex is in the region of self-intersection. Second, the partial derivatives could be easily derived by the normal vectors of neighboring triangles of the vertices. Finally, this penalty term could be applied in gradient-based optimization algorithms to remove the self-intersection of triangular meshes without using any approximation. Qualitative and quantitative evaluations were conducted to demonstrate the effectiveness and generality of our proposed method compared with previous approaches. The experimental results show that our proposed penalty term can avoid self-intersection to exclude unreasonable predictions and improves the accuracy of 3D pose estimation indirectly. Further more, the proposed method could be employed universally in triangular mesh based 3D reconstruction.


2017 ◽  
Vol 1657 ◽  
pp. 101-108 ◽  
Author(s):  
Julianna M. Kulik ◽  
Anthony P. Pawlak ◽  
Manraj Kalkat ◽  
Kevin R. Coffey ◽  
Mark O. West

2019 ◽  
Vol 3 (3) ◽  
pp. 471
Author(s):  
Tuong Thanh Nguyen ◽  
Van-Hung Le ◽  
Duy-Long Duong ◽  
Thanh-Cong Pham ◽  
Dung Le

Preserving, maintaining and teaching traditional martial arts are very important activities in social life. That helps preserve national culture, exercise and self-defense for practitioners. However, traditional martial arts have many different postures and activities of the body and body parts are diverse. The problem of estimating the actions of the human body still has many challenges, such as accuracy, obscurity, etc. In this paper, we survey several strong studies in the recent years for 3-D human pose estimation. Statistical tables have been compiled for years, typical results of these studies on the Human 3.6m dataset have been summarized. We also present a comparative study for 3-D human pose estimation based on the method that uses a single image. This study based on the methods that use the Convolutional Neural Network (CNN) for 2-D pose estimation, and then using 3-D pose library for mapping the 2-D results into the 3-D space. The CNNs model is trained on the benchmark datasets as MSCOCO Keypoints Challenge dataset [1], Human 3.6m [2], MPII dataset [3], LSP [4], [5], etc. We final publish the dataset of Vietnamese's traditional martial arts in Binh Dinh province for evaluating the 3-D human pose estimation. Quantitative results are presented and evaluated.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium provided the original work is properly cited.  


2017 ◽  
Vol 2 (3) ◽  

Melanoma is the most dangerous type of skin cancer in which mostly damaged unpaired DNA starts mutating abnormally and staged an unprecedented proliferation of epithelial skin to form a malignant tumor. In epidemics of skin, pigment-forming melanocytes of basal cells start depleting and form uneven black or brown moles. Melanoma can further spread all over the body parts and could become hard to detect. In USA Melanoma kills an estimated 10,130 people annually. This challenge can be succumbed by using the certain anti-cancer drug. In this study design, cyclophosphamide were used as a model drug. But it has own limitation like mild to moderate use may cause severe cytopenia, hemorrhagic cystitis, neutropenia, alopecia and GI disturbance. This is a promising challenge, which is caused due to the increasing in plasma drug concentration above therapeutic level and due to no rate limiting steps involved in formulation design. In this study, we tried to modify drug release up to threefold and extended the release of drug by preparing and designing niosome based topical gel. In the presence of Dichloromethane, Span60 and cholesterol, the initial niosomes were prepared using vacuum evaporator. The optimum percentage drug entrapment efficacy, zeta potential, particle size was found to be 72.16%, 6.19mV, 1.67µm.Prepared niosomes were further characterized using TEM analyzer. The optimum batch of niosomes was selected and incorporated into topical gel preparation. Cold inversion method and Poloxamer -188 and HPMC as core polymers, were used to prepare cyclophosphamide niosome based topical gel. The formula was designed using Design expert 7.0.0 software and Box-Behnken Design model was selected. Almost all the evaluation parameters were studied and reported. The MTT shows good % cell growth inhibition by prepared niosome based gel against of A375 cell line. The drug release was extended up to 20th hours. Further as per ICH Q1A (R2), guideline 6 month stability studies were performed. The results were satisfactory and indicating a good formulation approach design was achieved for Melanoma treatment.


Somatechnics ◽  
2015 ◽  
Vol 5 (1) ◽  
pp. 88-103 ◽  
Author(s):  
Kalindi Vora

This paper provides an analysis of how cultural notions of the body and kinship conveyed through Western medical technologies and practices in Assisted Reproductive Technologies (ART) bring together India's colonial history and its economic development through outsourcing, globalisation and instrumentalised notions of the reproductive body in transnational commercial surrogacy. Essential to this industry is the concept of the disembodied uterus that has arisen in scientific and medical practice, which allows for the logic of the ‘gestational carrier’ as a functional role in ART practices, and therefore in transnational medical fertility travel to India. Highlighting the instrumentalisation of the uterus as an alienable component of a body and subject – and therefore of women's bodies in surrogacy – helps elucidate some of the material and political stakes that accompany the growth of the fertility travel industry in India, where histories of privilege and difference converge. I conclude that the metaphors we use to structure our understanding of bodies and body parts impact how we imagine appropriate roles for people and their bodies in ways that are still deeply entangled with imperial histories of science, and these histories shape the contemporary disparities found in access to medical and legal protections among participants in transnational surrogacy arrangements.


2020 ◽  
Vol 2020 (17) ◽  
pp. 2-1-2-6
Author(s):  
Shih-Wei Sun ◽  
Ting-Chen Mou ◽  
Pao-Chi Chang

To improve the workout efficiency and to provide the body movement suggestions to users in a “smart gym” environment, we propose to use a depth camera for capturing a user’s body parts and mount multiple inertial sensors on the body parts of a user to generate deadlift behavior models generated by a recurrent neural network structure. The contribution of this paper is trifold: 1) The multimodal sensing signals obtained from multiple devices are fused for generating the deadlift behavior classifiers, 2) the recurrent neural network structure can analyze the information from the synchronized skeletal and inertial sensing data, and 3) a Vaplab dataset is generated for evaluating the deadlift behaviors recognizing capability in the proposed method.


Author(s):  
Anne Phillips

No one wants to be treated like an object, regarded as an item of property, or put up for sale. Yet many people frame personal autonomy in terms of self-ownership, representing themselves as property owners with the right to do as they wish with their bodies. Others do not use the language of property, but are similarly insistent on the rights of free individuals to decide for themselves whether to engage in commercial transactions for sex, reproduction, or organ sales. Drawing on analyses of rape, surrogacy, and markets in human organs, this book challenges notions of freedom based on ownership of our bodies and argues against the normalization of markets in bodily services and parts. The book explores the risks associated with metaphors of property and the reasons why the commodification of the body remains problematic. The book asks what is wrong with thinking of oneself as the owner of one's body? What is wrong with making our bodies available for rent or sale? What, if anything, is the difference between markets in sex, reproduction, or human body parts, and the other markets we commonly applaud? The book contends that body markets occupy the outer edges of a continuum that is, in some way, a feature of all labor markets. But it also emphasizes that we all have bodies, and considers the implications of this otherwise banal fact for equality. Bodies remind us of shared vulnerability, alerting us to the common experience of living as embodied beings in the same world. Examining the complex issue of body exceptionalism, the book demonstrates that treating the body as property makes human equality harder to comprehend.


Author(s):  
Rajendra Pai N. ◽  
U. Govindaraju

Ayurveda in its principle has given importance to individualistic approach rather than generalize. Application of this examination can be clearly seem like even though two patients suffering from same disease, the treatment modality may change depending upon the results of Dashvidha Pariksha. Prakruti and Pramana both used in Dashvidha Pariksha. Both determine the health of the individual and Bala (strength) of Rogi (Patient). Ayurveda followed Swa-angula Pramana as the unit of measurement for measuring the different parts of the body which is prime step assessing patient before treatment. Sushruta and Charaka had stated different Angula Pramana of each Pratyanga (body parts). Specificity is the characteristic property of Swa-angula Pramana. This can be applicable in present era for example artificial limbs. A scientific research includes collection, compilation, analysis and lastly scrutiny of entire findings to arrive at a conclusion. Study of Pramana and its relation with Prakruti was conducted in 1000 volunteers using Prakruti Parkishan proforma with an objective of evaluation of Anguli Pramana in various Prakriti. It was observed co-relating Pramana in each Prakruti and Granthokta Pramana that there is no vast difference in measurement of head, upper limb and lower limb. The observational study shows closer relation of features with classical texts.


Author(s):  
Brandon Shaw

Romeo’s well-known excuse that he cannot dance because he has soles of lead is demonstrative of the autonomous volitional quality Shakespeare ascribes to body parts, his utilization of humoral somatic psychology, and the horizontally divided body according to early modern dance practice and theory. This chapter considers the autonomy of and disagreement between the body parts and the unruliness of the humors within Shakespeare’s dramas, particularly Romeo and Juliet. An understanding of the body as a house of conflicting parts can be applied to the feet of the dancing body in early modern times, as is evinced not only by literary texts, but dance manuals as well. The visuality dominating the dance floor provided opportunity for social advancement as well as ridicule, as contemporary sources document. Dance practice is compared with early modern swordplay in their shared approaches to the training and social significance of bodily proportion and rhythm.


Sign in / Sign up

Export Citation Format

Share Document