search

wdhudiekou / UMF-CMGR

[IJCAI2022 Oral] Unsupervised Misaligned Infrared and Visible Image Fusion via Cross-Modality Image Generation and Registration
MIT License
153 stars 17 forks source link

Qusetion about raw dataset #10

Closed socome closed 2 years ago

socome commented 2 years ago

Thank you for great your work!

I have a question about your datasets.

I get the RoadScene dataset from here, and then there are two types of thermal images(cropinfrared and infrared).

What is correct infrared image for making the test images using this code..?

To my understanding, I use the corp infrared images for making the test dataset. Is it right?

Thank you. I'm waiting your reply.

socome commented 2 years ago

In addition, Train/Test data are not clear.. How can I get the splitted RoadScene datasets to reproduce your paper. Could you provide the pretrained weight for testing your model...?

wdhudiekou commented 2 years ago

The raw images of RoadScene are large in size. To speed up training, we resize training and testing images to 256*256 in this code. You can also adjust these images to other sizes. In addition, the pretrained model will be released as soon as possible.

socome commented 2 years ago

@wdhudiekou Thank you for your answer. However, My question is certainly not resolved.

Q1. I get the RoadScene dataset from here, and then there are two types of thermal images(cropinfrared and infrared). To my understanding, I use the corp infrared images for making the test dataset. Is it right?

Q2. How can I get the splitted RoadScene datasets to reproduce your paper.

Thank you. I'm waiting your reply.

socome commented 2 years ago

@wdhudiekou I'm waiting your reply.

wdhudiekou commented 2 years ago

For making the test images, you can use the crop infrared images and get_test_data.py