Open Koteswara-ML opened 4 years ago
ResNet is the CNN used to generate descriptors for the bounding box images. To use grayscale images you will have to retrain the CNN if ResNet does not suffice as is. You may be able to use ResNet by converting the grayscale images to RGB as mentioned here: https://stackoverflow.com/questions/51995977/how-can-i-use-a-pre-trained-neural-network-with-grayscale-images
What all enhancements can be done to actually improve the performance of the tracking the objects in gray-scale videos..?
can some one help me with this..