lyxbyr
lyxbyr
same issue,could you solve it?
Thanks for your prompt reply, I'm sorry I didn't make it clear amodel_offset is offset (∆ u , ∆ v ) to the keypoint location ? In this way, because...
head:hm value need to set correct,and coco category value could not 0 ,need to 1.2...
batch_size,lr need to finetune acoording gpu numbers, can times relationship ，another in author model finetune use --load_model
i set my dataset in data folder like coco format, not using --imag_path --annotation_path, python main.py tracking --exp_id coco_tracking --tracking --gpus 0 --batch_size 8 --lr 6e-5 --num_workers 4 --pre_hm --shift...
try to edit coco.py review num_categories = your class number ,class_name = ['your class'], _valid_ids=[1,2,3,and so on]? At 2021-01-10 15:46:39, "Hidehidden" wrote: Hello,I have always been concerned about your outstanding...
@Cknex Hi ， similar performance how to eval it ? computer dimension error and orientation error ？the final three branch fc layer node need to change,eg resnet18 or resnet50 the...
**my result: Angle error Dimension error Vgg19 13.683045 0.283335 resnet34 27.887893 0.849132 resnet50 21.706590 0.821558**
Hi,for resnet50,fc like this result may be better : self.orientation = nn.Sequential( nn.Linear(2048 * 7 * 7, 1024), nn.ReLU(True), nn.Dropout(), nn.Linear(1024, 1024), nn.ReLU(True), nn.Dropout(), nn.Linear(1024, bins * 2) # to...