Training Details (Skip this part if you arent interested in AI programming) - I used 2000 synthetic 512x512 px training images (plus 200 synthetic validation images) - I did ZERO training on real images - I trained the "heads" for 4 epochs of 500 steps and the full network for another 4 epochs - Resnet50 backbone (Resnet101 caused OOM errors) using COCO pre-trained weights - Training took about 20 minutes on my computer - Final losses after the 8th epoch: loss: 0.3615 - rpn_class_loss: 0.0023 - rpn_bbox_loss: 0.1823 - mrcnn_class_loss: 0.0232 - mrcnn_bbox_loss: 0.0636 - mrcnn_mask_loss: 0.0901 - val_loss: 0.3261 - val_rpn_class_loss: 0.0013 - val_rpn_bbox_loss: 0.1875 - val_mrcnn_class_loss: 0.0043 - val_mrcnn_bbox_loss: 0.0535 - val_mrcnn_mask_loss: 0.0794 Results It's far from perfect, and probably isn't practical to bolt onto a robot right now, but with more training and improvements in GPU hardware, I think this could be a viable solution to pick up cigarette butts on a huge scale
New requirements for signage on ashtrays in no-smoking areas
Leaders in smaller jurisdictions might say, Yes but we are not New York City or Minneapolis. At that point, police chiefs need to be blunt with their mayors saying something like, Respectfully your honor, we may not be plagued by as many shootings as New York City and Minneapolis, however defunding will lead to a precipitous drop in all enforcement activities
"Lollapalooza 2016 Will Last 4 Days, Cost 20% More"
To all pipe smokers, avoid large pipes at all costs
