Machine Learning Engineer (II)
Paid internship at FotoNation · Starts after academic term
City:
  • room București

Xperi invents, develops and delivers technologies that enable extraordinary experiences. We make entertainment more entertaining, and smart devices smarter. Content markets are changing, fast. Consumers face a simultaneously expanding and fragmenting set of choices. Xperi technology cuts through the chaos, putting us at the forefront of fast-moving trends in streaming, digital entertainment, and AI applications in any environment. Working from home was never more common than these days of pandemic and lockdowns. We all stay connected with video conferencing, and, with obvious benefits of feeling closer to our colleagues we get all the drawbacks of exposing not-so-glamorous aspects of our work-from-home lives. In this situation, many turned to virtual background feature to create a more professional environment. However, even with popular video conference platforms, virtual background implementations are not showing particularly high quality. In line with the recent success of convolutional neural networks (CNNs) for semantic segmentation, we develop human body segmentation methods and segmentation labelling tools for mobile applications and conferencing. Your job will be to study and experiment with human body segmentation, background subtraction and image matting solutions.

Skills required:

  • Ability to write robust code in Python
  • Outstanding analytical and problem-solving skills
  • Deep knowledge of math, probability, statistics and algorithms
  • Familiarity with machine learning frameworks (PyTorch) and libraries

Nice to have:

  • BSc in Computer Science, Mathematics or similar field

References:

  • Shanchuan Lin et al., “Real-Time High-Resolution Background Matting”, arXiv, 2020
  • Hierarchical Multi-Scale Attention for Semantic Segmentation, Andrew Tao, Karan Sapra, Bryan Catanzaro, arXiv, 2020