poppinjie

我在工程中看到一个文件夹保存有一些主流跟踪器在OTB2015上测试的结果，请问作者有这些跟踪器在OTB2013上测试的结果吗？能提供一下？万分感谢！

What the meaning of the 'response_up=8' in tracking.py?

2

I am confused with the viriable 'response_up' in 'disp_in_xcrop = disp_in_area * float(tot_stride) / response_up',in addition,what is the reason for calculating the distance like this.Can you explain it?

the effect of the 'weight' in the function 'F.binary_cross_entropy_with_logits'

2

Hello!Thanks for your work.It's complete(including training and tracking) and perfect. When I read the training code, I had a question. What's the effect of 'weight ' in the call 'F.binary_cross_entropy_with_logits'...

Why do you use cv2.warpAffine to attain the goal of cropping an image rather than crop the image derectly?

I see you use 'np.roll()' function when you define gaussian_shaped_labels in the util.py,why you do that? It seems not accord with the definition of gaussian label.I hope you can help...

How to understand the values 0.625, 1.25, 2.5, 5.0 in the PWCNet.py

6

I find that these values are multiplied by the upsampled optical flow.But I am not clear what's the meanning of these operations.Why the upsampled optical flow need to multiply corresponding...

What's does the values [0.45, 0.432, 0.411] mean?

2

In the main.py file,there is a Normalization operation which is " transforms.Normalize(mean=[0.45, 0.432, 0.411], std=[1, 1, 1])". I was confused of the meaning of the values [0.45, 0.432, 0.411]. I...

How to understand the "training for multiple flow" and the "div_flow"

1

div_flow is used for 2 reasons : Training is deemed more stable for low output target values. it was used by initial caffe code, and thus it makes this training...

NameError: name 'resnet_v1_101_flownet_rfcn' is not defined

1

When I run the "dff_rfcn/demo.py", I encounted this error. Anybody know why? Please help me out!

我想问下关键帧t到t-K的光流方向与t到t+K的光流方向是相反的,那么在做warp的时候会有什么不同吗?而且用flownet训练的是正向光流的估计,那么t到t-K的光流估计效果是不是不太好?