the depth_conf is all nearly 1.0

[GuitarYourself]

it seems that the depth_conf result of the model is all nearly 1.0， is that correct？

[jytime]

Hi,

1. Can you check the max and median distribution of the depth conf map?
2. If the max is still around 1.0, it is quite possible that (a) a lot of dynamic pixels or (b) really little overlap
3. You can check our examples (under vggt/examples) to check what's the normal distribution of depth map conf.

[GuitarYourself]

Hi,

1. Can you check the max and median distribution of the depth conf map?
2. If the max is still around 1.0, it is quite possible that (a) a lot of dynamic pixels or (b) really little overlap
3. You can check our examples (under vggt/examples) to check what's the normal distribution of depth map conf.

thx for your reply, the max is around 1.4, about half is greater than 1.0. maybe my images have little overlap, I will try more data. Thx.

[xuyankun]

@jytime Hi, I am wondering generated confidence_map is >1, but your demo on Hugging face is a filtering bar ranged in [0, 100], how do you convert this unit?

[jytime]

Hey! The filtering bar in the HF demo represents a percentage value. We scale the confidence map to [0, 100] for visualization.