LeslieCasas

Chair for Computer Aided Medical Procedures & Augmented Reality
Lehrstuhl für Informatikanwendungen in der Medizin & Augmented Reality

THIS WEBPAGE IS DEPRECATED - please visit our new website

Leslie Casas, M.Sc.

leslie.casas@tum.de

Chair for Computer Aided Medical Procedures & Augmented Reality
Fakultät für Informatik / I-16
Technische Universität München
Boltzmannstraße 3
85748 Garching bei München
Germany

Raum MI 03.013.037
Telefon CAMP: +49 (89) 289-17077

this is me

Research Interests

  • Human Pose Estimation
  • Human gait analysis
  • Inertial Sensors
  • Machine Learning

Active research projects

Patient Pose Estimation

Patient Pose Estimation

In this project, we investigate machine learning algorithms for the purpose of real-time 3D human pose estimation in a hospital monitoring scenario. Patients are typically covered with a blanket and the bed is cluttered with books, laptops and other everyday objects. This makes the task more challenging than background-free human pose estimation. With the constraint of day and night monitoring capacity, we have recently published a method based on depth data which will be presented at MICCAI 2016, see link to the paper below. On this project page, you can also find the used dataset and training code. For further information, please contact Felix Achilles.
Adversarial Learning for Denoising

Adversarial Learning for Denoising

We present the idea of adversarial learning for one-dimensional signal denoising. An adversarial encoder-decoder network is proposed for denoising signals, represented by a sequence of measurements. We rely on the network’s latent representation as input to the discriminator to classify the signal into clean or noisy. Aligning the noisy and clean signal distributions results in removing the signal noise. Unlike the standard GAN training, we propose a new formulation that suits to one-dimensional signal denoising. In the evaluation, we show better performance than the related approaches, such as autoencoders and recurrent neural networks, demonstrating the benefits of our adversarial encoder-decoder network in different signal and noise types.

Student projects

We continuously create new topics for student projects or bachelor- and master-theses.
Feel free to contact me, if you are interested in one of the topics seen here.
  • Human Pose from Inertial sensors
  • Inertial Sensor Fusion

Publications

2018
L. Casas, N. Navab, V. Belagiannis
Adversarial Signal Denoising with Encoder-Decoder Networks
This works can be found at: https://arxiv.org/abs/1812.08555 (bib)
L. Casas, N. Navab, S. Demirci
Patient 3D body pose estimation from pressure imaging
International Journal of Computer Assisted Radiology and Surgery, https://doi.org/10.1007/s11548-018-1895-3 (bib)
L. Casas, C. Muerwald, F. Achilles, D. Mateus, D. Huber, N. Navab, S. Demirci
Human Pose Estimation from Pressure Sensor Data
Bildverarbeitung fuer die Medizin 2018, pp 285-290 (bib)


UsersForm
Title: M.Sc.
Circumference of your head (in cm):  
Firstname: Leslie
Middlename:  
Lastname: Casas
Picture: LeslieCasas.png
Birthday:  
Nationality: Blank
Languages:  
Groups: Patient Monitoring, Computer Vision
Expertise:  
Position: Scientific Staff
Status: Active
Emailbefore: leslie.casas
Emailafter: tum.de
Room: MI 03.13.037
Telephone: +49 (89) 289-17077
Alumniactivity:  
Defensedate:  
Thesistitle:  
Alumnihomepage:  
Personalvideo01:  
Personalvideotext01:  
Personalvideopreview01:  
Personalvideo02:  
Personalvideotext02:  
Personalvideopreview02:  


Edit | Attach | Refresh | Diffs | More | Revision r1.12 - 30 Jan 2019 - 17:55 - LeslieCasas

Lehrstuhl für Computer Aided Medical Procedures & Augmented Reality    rss.gif