search

htcr / sam_road

Segment Anything Model for large-scale, vectorized road network extraction from aerial imagery. CVPRW 2024
https://arxiv.org/pdf/2403.16051.pdf
MIT License
138 stars 18 forks source link

How to set the parameters in a YAML file(eg : toponet_vitb_512_cityscale.yaml) for testing my own dataset? #23

Open xiaofei-guan opened 3 months ago

xiaofei-guan commented 3 months ago

I successfully ran inferencer.py to test my own dataset, which includes images collected from various online maps. Two sample images from the dataset are displayed below. To align with the images in Cityscale, I resized all my images to 2024x2024 pixels. However, I observed poor results, suggesting the algorithm did not perform well. One sample images from the results is displayed below. I guess that adjusting the YAML parameters appropriately might improve the outcome. Could you please provide a more detailed explanation of these parameters and their functions?

0

htcr commented 3 months ago

Hi, first of all, I think you may want to try to scale the images such that they are 1.0m / pixel. That's how these checkpoints are trained. Can you give this a try first?