DCFNet_pytorch
DCFNet_pytorch copied to clipboard
foolwood
DCFNet: Discriminant Correlation Filters Network for Visual Tracking
Hey, Is there any newer implementation of the Discriminant Correlation Filters, which are working with the newer pytorch versions? They changed a lot at their complex number calculations, so the...
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...
Thank you for your excellent work and your share! These days,I copy the code and retrain the DCFnet, to my surprise,I find that although the mean loss is decreasing, the...
Randomly select [search_id](https://github.com/foolwood/DCFNet_pytorch/blob/master/train/dataset.py#L25). Will there be cases where target and search are different objects?
Parallelising the image crop for faster preparation of the dataset for training
By keeping your default parameters, without fine-tune, I got 0.6466 on OTB2013 and 0.6206 on OTB2015 (guess you are using py2 so I also tested using py2). So is that...
When the data is preprocessed, the images containing the occluded and incomplete tracking objects are removed, yes? How does this affect the training effect?
when I run the command for training, the cuFFT error happened. It happened in the line 47 of net.py and it has a tips of "RuntimeError: cuFFT error: CUFFT_ALLOC_FAILED "....
