What if I want to detect multiple objects in a single image. What should be the approach?
Image classification is different from object detection and segmentation.
The models and datasets used are very different as well.
If you are interested in object detection check out the COCO dataset, and YOLO / DarkNet model
Thanks rehoprincewill for your response!
I understand that object detection is quite different from image classification! But what I want to do is something like this: I have some 10 classes/types of species in an image (It’s a 3D image with x,y,z dimensions actually, with two numerical values that classify the characteristics of the object in that pixel). Suppose I want to classify such species in an 3D image, what technique should I use. I have to label the species, just to be clear! So will that still be considered as object detection?
Yes it will.
To be considered classification, only a species/class should be present in the input.