search

gabrieleilertsen / hdrcnn

HDR image reconstruction from a single exposure using deep CNNs
https://computergraphics.on.liu.se/hdrcnn/
BSD 3-Clause "New" or "Revised" License
502 stars 101 forks source link

Using the encoder for anomaly detection #23

Closed Tamoghna-Saha closed 4 years ago

Tamoghna-Saha commented 5 years ago

Hi,

I have run your model on multiple different images and it has given me perfect reconstruction every time. This led me to believe that the encoder model can be used directly for anomaly detection.

Is it possible for me to acquire the encoder weights for me to test this hypothesis? Additionally, how should I modify your current codes with respect to the shape of the input/output and the exact line of codes that will require modification, etc?

Please let me know if you have any other ideas/hypothesis.

gabrieleilertsen commented 5 years ago

I am not sure I understand what you mean by "anomaly detection".

The weights are available for download. Setting up the model and importing the weights as in the prediction script, you can explore the weights or activations from each layer of the encoder for a certain image.