Deep-homography-estimation-Pytorch
Deep-homography-estimation-Pytorch copied to clipboard
Deep homography network with Pytorch
Hi mazenmel, I have a question for you. save .npy files function in DataGenerationAndPRocessing.py np.save(new_path+'%s'%i[0:12], ImagePreProcessing(i,path)) ImagePreProcessing(i,path) is a tuple. cuz return value of ImagePreProcessing is Tuple (variable is named...
TrainingData = CocoDdataset(train_path)This line of code has gone wrong.
array = np.load(path+'%s'%i)
Thanks for re-implement the model. How's the performance? Consistent with claimed in the paper?
how do I infer on a single pair of image?