Determining objects in a video is much easier for a deep learning network than you think. In this video, learn how to classify the objects in a video using the GoogLeNet model.
- [Instructor] So let's get started with using OpenCV's … dnn module as an inference engine for a video file. … Now, earlier in this course we looked at working … with video files with OpenCV, so let's use the file … that we created there as a starting point. … We've also used OpenCV for image classification, and so … we can reuse much of this for working with video files. … Now, we're going to be using the same shore video, … and let's head over to the image classification file. … And we've worked with these synset files … for the 1000 classes, so let's reuse that, … along with the list comprehension, and let's load … our caffe files, so I'm going to grab the three rows. … So that's all_rows, the classes, and reading the files … into caffe, and I'm going to copy it into my video file. … Now, the next thing we do is to confirm … that we can actually open the video file or stream. … Now, a video is made up of multiple images; … so all we're doing is taking each image … of frame and checking against the model. …
- Deep learning for OpenCV
- Viewing images and video in OpenCV
- Working with blobs in the dnn module
- Image classification
- Video classification