PerceptualImageError
PerceptualImageError copied to clipboard
prashnani
A metric for Perceptual Image-Error Assessment through Pairwise Preference (PieAPP at CVPR 2018).
Hi, I have tested the trained model(download_PieAPPv0.1_PT_weights.sh) that you released, but the performance is lower than the result in your paper by about 2% in both PLCC, SRCC, and KRCC....
Even with --gpu_id 0 only CPU is used when using the windows executable, any workaround for that?
Sorry for my noob question, but Is possible train this network with my dataset (images from one class) and get same results???
Tested to work with Python 3 and the version of PyTorch currently installed by conda (0.4.1). Also, added `.gitignore` to ignore files generated by Python that shouldn't be committed and...
Hi! Will the training code be uploaded any time soon? Am excited to try and train with this metric.
 how can i use numpy in pytorch?
Below is the error message: "size mismatch for ref_score_subtract.weight: copying a param with shape torch.Size([1]) from checkpoint, the shape in current model is torch.Size([1, 1])." Need to manually adjust the...
The two files contain code for implementation in keras to the PIE network. The file on the model folder contain a class PieAPP that builds the model, load weights from...
Simple changes in the code to make it work with more recent version of python (tested with python 3.11.5 and pytorch 2.0.1+cu118)
Hi,i tried to simulate the geometric transformation in distorted images,but i don't understand the method that the hole-filling is done using an open-source tool GMIC,does it means Fill Holes in...
