Reviewing Mask R-CNN: an In-Depth Analysis of Models and Applications

EasyChair Preprint no. 11838

Abstract

This comprehensive review delves into the intricate realm of Mask R-CNN, conducting a meticulous analysis of its various models and applications within the field of computer vision. Mask R-CNN, known for its prowess in instance segmentation, is dissected in terms of architecture, design, and performance metrics. The review explores its diverse applications, ranging from image and video segmentation to medical image analysis and autonomous driving. Emphasizing the importance of representative datasets, the training process is elucidated, encompassing data preprocessing and model optimization techniques. Strengths such as accuracy in instance segmentation and versatility in handling different object scales are highlighted, along with a discussion of limitations and challenges. A comparative analysis with other state-of-the-art models offers insights into Mask R-CNN's relative strengths and weaknesses. The review concludes by outlining future research directions and the model's potential contributions to the evolution of computer vision applications.

Keyphrases: Applications, Mask R-CNN, models

@Booklet{EasyChair:11838,
  author = {Karem Mohammed},
  title = {Reviewing Mask R-CNN: an In-Depth Analysis of Models and Applications},
  howpublished = {EasyChair Preprint no. 11838},

  year = {EasyChair, 2024}}