Computer Vision Group, RWTH Aachen University

  • Developed a novel and efficient architecture (DynaMITe) for multi-instance interactive object segmentation using Transformer.
  • Achieved state-of-the-art results on multiple existing interactive segmentation benchmarks and the proposed new multi-instance benchmark.
  • Gained expertise in efficient large-scale multi-GPU training of deep learning models across multiple nodes.
  • HTML5
  • CSS3
  • JavaScript
  • jQuery
  • Wordpress