Blobs are the tensors of OpenCV. In this video, learn how to ensure they have the required shapes for use in a deep learning network.
- [Instructor] Now a blob is one or more images … with the same width, height, and number of channels … that have all been preprocessed in the same way. … Let's take a look at the workflow … for an inference engine and where blobFromImage fits in. … What we will do is pass one image to blobFromImage. … Now if you have more than one image, … you'll use the blobFromImages function. … The output of this is a four dimensional tensor … where N is the number of images, … C is the number of channels, … H is the height of the tensor, … and W is the width of the tensor. … This is stored in a blob object. … This is then passed to a trained model … that allows us to get the image or video inference. … For images and videos, this could be object detection, … image classification, semantic segmentation, … and a whole load more. … It depends on what trained model you have used … and your use case. … Let's take a look at the parameters for blobFromImage. … BlobFromImage creates a four dimensional blob from an image …
- Deep learning for OpenCV
- Viewing images and video in OpenCV
- Working with blobs in the dnn module
- Image classification
- Video classification