Filters








11 Hits in 6.5 sec

Articulated clinician detection using 3D pictorial structures on RGB-D data

Abdolrahim Kadkhodamohammadi, Afshin Gangi, Michel de Mathelin, Nicolas Padoy
2017 Medical Image Analysis  
Instead, we propose a novel approach based on Pictorial Structures (PS) and on RGB-D data, which can be easily deployed in real ORs. We extend the PS framework in two ways.  ...  Our approach is evaluated for pose estimation and clinician detection on a challenging RGB-D dataset recorded in a busy operating room during live surgeries.  ...  Index Terms-Medical computer vision, surgical activity analysis, clinician pose estimation, 3D pictorial structures, RGB-D Data I.  ... 
doi:10.1016/j.media.2016.07.001 pmid:27449279 fatcat:t5hugb43bjg5dhkg26cr2fkaze

A Multi-view RGB-D Approach for Human Pose Estimation in Operating Rooms [article]

Abdolrahim Kadkhodamohammadi, Afshin Gangi, Michel de Mathelin, Nicolas Padoy
2017 arXiv   pre-print
We evaluate this approach on a novel multi-view RGB-D dataset acquired during live surgeries and annotated with ground truth 3D poses.  ...  In this paper, we propose an approach for multi-view 3D human pose estimation from RGB-D images and demonstrate the benefits of using the additional depth channel for pose refinement beyond its use for  ...  First, we extend the 3D pictorial structures of [23] to use a ConvNet body part detector on RGB-D images.  ... 
arXiv:1701.07372v1 fatcat:vztmw5hr4fgdlljqsaffcbpvs4

Unsupervised domain adaptation for clinician pose estimation and instance segmentation in the operating room [article]

Vinkle Srivastav, Afshin Gangi, Nicolas Padoy
2022 arXiv   pre-print
We propose to exploit explicit geometric constraints on the different augmentations of the unlabeled target domain image to generate accurate pseudo labels and use these pseudo labels to train the model  ...  The fine-grained localization of clinicians in the operating room (OR) is a key component to design the new generation of OR support systems.  ...  RGB-D extensions to the pictorial structure model.  ... 
arXiv:2108.11801v4 fatcat:4fxsoswfpvfejbcqmzvkdorpky

Optimizing Nondecomposable Loss Functions in Structured Prediction

Mani Ranjbar, Tian Lan, Yang Wang, Steven N. Robinovitch, Ze-Nian Li, Greg Mori
2013 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We show significant improvement over baseline approaches that either use simple loss functions or simple scoring functions on the PASCAL VOC and H3D Segmentation datasets, and a nursing home action recognition  ...  We develop an algorithm for structured prediction with nondecomposable performance measures.  ...  Fig. 3 .3(c,d) illustrates the probabilities for pictorial structure (generic edge templates) and iterative parsing models (after color information).  ... 
doi:10.1109/tpami.2012.168 pmid:22868650 pmcid:PMC3547074 fatcat:etkqq47cjfctdd4cmnrkgddshe

Unsupervised domain adaptation for clinician pose estimation and instance segmentation in the OR [article]

Vinkle Srivastav, Afshin Gangi, Nicolas Padoy
2021
We propose to exploit explicit geometric constraints on the different augmentations of the unlabeled target domain image to generate accurate pseudo labels, and using these pseudo labels to train the model  ...  The fine-grained localization of clinicians in the operating room (OR) is a key component to design the new generation of OR support systems.  ...  RGB-D extensions to the pictorial structure model.  ... 
doi:10.48550/arxiv.2108.11801 fatcat:sfmrlp46qvbxxbi4guktv7g6pm

Comparative Analysis of Different Machine Learning Classifiers for the Prediction of Chronic Diseases [chapter]

Rajesh Singh, Anita Gehlot, Dharam Buddhi
2022 Comparative Analysis of Different Machine Learning Classifiers for the Prediction of Chronic Diseases  
Precise diagnosis of these diseases on time is very significant for maintaining a healthy life.  ...  Chronic Diseases are the most dangerous diseases for humans and have significant effects on human life. Chronic Diseases like heart disease & Diabetes are the main causes of death.  ...  The region-based segmentation will segment the data dependent on the taken-out features using GLCM algorithm.  ... 
doi:10.13052/rp-9788770227667 fatcat:da47mjbbyzfwnbpde7rgbrlppe

Markerless Human Motion Analysis

MATTEO MORO
2022
from RGB video data.  ...  State-of-the-art technologies that provide useful and accurate quantitative measures rely on marker-based systems.  ...  Figure 3 . 3 : 33 Figure 3.3: Examples of body models representing stick figures (a) and pictorial structures (b-c).  ... 
doi:10.15167/moro-matteo_phd2022-04-22 fatcat:o6zubebmwnf5zfefrs2jru7z7q

Final Program, The International Neuropsychological Society, The Polish Neuropsychological Society and The Polish Neuroscience Society Joint Mid-Year Meeting

2010 Journal of the International Neuropsychological Society  
For each subject, an axial 3D T1 acquisition was obtained with the following parameters: TR=10.3ms, TE=10ms FOV=26mm, matrix size=320×224, 152 slices. The anatomical 3D data were analyzed with SPM5.  ...  In other experiment different manipulation was used, namely R and B components of RGB color space were shifted.  ...  C Co on nc cl lu us si io on ns s: : These data suggest that normal variations in vascular risk factors have specific structural consequences on brain structures in otherwise healthy adults.  ... 
doi:10.1017/s1355617710001062 fatcat:kru7xx2k3zeazneuyevxrov3cu

Multimodal Emotion Recognition Based Human-Robot Interaction Enhancement

Fatemeh Noroozi
2020 unpublished
Thus, on the synchronized data modalities, we applied face detection using the method of [195] on RGB modality and cropped associated faces on D and T modalities by using computed homographs.  ...  The detections can then be refined according to the structure of human body. Grammar models and pictorial structures are some examples.  ...  KOKKUVÕTE (SUMMARY IN ESTONIAN) MULTIMODAALSEL EMOTSIOONIDE TUVASTAMISEL PÕHINEVA INIMESE-ROBOTI SUHTLUSE ARENDAMINE Üks afektiivse arvutiteaduse peamistest huviobjektidest on mitmemodaalne emotsioonituvastus  ... 
doi:10.13140/rg.2.2.17543.96164 fatcat:sqfxpplvgndwjhgkci7f5uexny

PROCEEDINGS OF THE 15TH PYTHON IN SCIENCE CONFERENCE SCHOLARSHIP RECIPIENTS JUMP TRADING AND NUMFOCUS DIVERSITY SCHOLARSHIP RECIPIENTS

Katy Huff, David Lippa, Dillon Niederhut, M Pacer, Serge Rey, Program Chairs, Juan Communications, Shishido, David Lippa, Katy Huff, M Pacer, Dillon Niederhut (+44 others)
Proceedings of the 15th Python in Science Conference July 10-July 16 •   unpublished
Francesco Pontiggia for helping us solidify many of our data handling and computing ideas and Dr. Jian-You Lin for being the first independent tester of our toolchain.  ...  The program was developed to analyze the results of simulations of phase change materials carried out on supercomputers in the Forschungszentrum Jülich.  ...  This library includes algorithms and data structures for both one-to-one and many-to-one matching.  ... 
fatcat:pkgs32mdl5d3tbfa5xkveedypq

Guidelines to assist building effective educational applications and e-games for children with ADHD

Doaa Sinnari
2019
Significant improvements found on their cognition, behaviour, social skills and academic performance.  ...  The most significant and effective methods/features from the included studies were highlighted and used to draw out our list of guidelines.  ...  It is a pictorial character that aids the process of merging emotions while texting using different messaging applications (Pohl et al., 2017) .  ... 
doi:10.15126/thesis.00850159 fatcat:asibigtvprdatd46gyx6ebeu3u