i've been doing some work on object detection and have looked at the config file of ssd_mobilenet_v2_quantized_300x300_coco.
my understanding is that the single shot multibox detector developed consists of a VGG16 base network with SSD layers attached. my question would be where do i see this architecture now and why does it not show up in the file.
I am only interested in the understanding that I would like to have before the training.
google-ml-butler[bot]
commented
1 year ago
hey guys,
i've been doing some work on object detection and have looked at the config file of ssd_mobilenet_v2_quantized_300x300_coco.
my understanding is that the single shot multibox detector developed consists of a VGG16 base network with SSD layers attached. my question would be where do i see this architecture now and why does it not show up in the file.
I am only interested in the understanding that I would like to have before the training.