How does maskrcnn work?

As a curious observer in the realm of computer vision and machine learning, I'm intrigued by the capabilities of Mask R-CNN, a state-of-the-art framework for object detection and segmentation. Could you elaborate on how Mask R-CNN works? Specifically, I'm interested in understanding the key components and mechanisms that enable it to identify and localize objects in an image, while also generating pixel-level masks for each detected object. Additionally, I'd like to know how Mask R-CNN builds upon its predecessor, Faster R-CNN, and what improvements it brings to the field of object detection and segmentation.