Haoyu Ma

Thanks for your asking. That's an interesting question. Personally, you can still do it based on a Stacked Hourglass Network. In detail, you can replace each hourglass module to the...

> That's a good one, but the parameters and computation of the stacked HR-Net will increase significantly compared to the original HR-Net. What do you think？ You are right. However,...

> Decreasing the number of blocks is indeed an effective way to reduce the number of parameters in one module.But decreasing the branch will also remove the low-resolution representation, I...

> Thank you very much for your suggestion, I think it is feasible. Excuse me for your precious time, thanks again. Thanks for proposing this interesting question. Actually, I haven't...

Sorry for the late reply. It is possible. As we only consider the synthetic mask loss as an auxiliary loss, our ultimate goal is the keypoint loss. After 50 epochs,...

Hi, you can change the label here. By default, 0 is for the "person" class. https://github.com/HowieMa/DeepSORT_YOLOv5_Pytorch/blob/master/main.py#L248 Note that, since the embedding is only for "person", you may also need to...

> Hi @HowieMa @DimaMirana , > > If I want to use the DeepSORT to detect only vehicles classes from the coco dataset (3-car, 4-motorbike, 6-bus and 8-truck for example)...

> Hi @HowieMa , > > Thanks for the clarification and reference, so just to confirm, the embedding is similar to a deep learning model (eg .pt) but instead trained...

Thanks for your asking. I am sorry for the ambiguous variable names. Nevertheless, the name of variables does not affect the results of our paper. Since we aim to build...

Hi, thanks for asking. Could you please show the specific results you get, rather than just say "higher"? I didn't try Resnet34 on CIFAR-100, but I did try ResNet18 and...