This would allow you to handle pixel-wise labels. Thus, the underlying objects can be of any shape. It is used for general semantic segmentation tasks.
I am assuming pixel labelling here means to doing Semantic Segmentation of the image you may be working on. SincercnnObjectDetectoris used specifically for Object Detection inaimage and Object classification it cannot be used for doing semantic segmentation. You may get started withSemantic Segmentation using Deep Learning for pixel labelling.
For learning more about semantic segmentation you may refer to the following:
Best Answer