R-CNN in depth

In this video, we will try to understand each and every aspect of the RCNN architecture.
RCNN is an object detection algorithm that consists of the four major parts:
1. Generating Region Proposals.
2. Finetuning the CNN network (AlexNet) and extracting features.
3. Training class wise SVMs.
4. Training the bounding box regressor.
For generating the region proposal we use the Selective Search algorithm.
RCNN forms the basis of Fast RCNN, Faster RCNN and Masked RCNN.
For a 1-1 discussion on your projects, take my Fiverr gig:
www.fiverr.com/share/94Pk9a
Instagram:
/ datumlearning