Indoor environment

[suyuzhang]

Hi @mdfirman,
Thanks for your work!

Does this method work in the indoor environment?

Thanks!

[agenthong]

Actually, you can try this.

[wqx854987945]

@suyuzhang Hi，have you tried this method in the indoor environment？

[agenthong]

Hi!

I train this for indoor scenes. But get depth a little bit weird:

I'm confused is there any mistake I made?

[ChauChorHim]

Hi!

I train this for indoor scenes. But get depth a little bit weird:

I'm confused is there any mistake I made?

Have you changed the max_depth and min_depth in the options.py. It's better if you can share your setting.

[agenthong]

Hi!
I train this for indoor scenes. But get depth a little bit weird:
I'm confused is there any mistake I made?

Have you changed the max_depth and min_depth in the options.py. It's better if you can share your setting.

Hi!

I only change the max_depth to 10 in options.py, and the min_depth is 0.1. So what else should I pull the change(like some max_depth settings in training code)?

[LeeBY68]

Hi!
I train this for indoor scenes. But get depth a little bit weird:
I'm confused is there any mistake I made?

Have you changed the max_depth and min_depth in the options.py. It's better if you can share your setting.

Hi!

I only change the max_depth to 10 in options.py, and the min_depth is 0.1. So what else should I pull the change(like some max_depth settings in training code)?

Which indoor dataset do you use?
The results are the predicted depth using the KITTI trained model? Or you train the model in the indoor dataset, the final results are these?

[daniyar-niantic]

Hi. Have you changed intrinsics?

[lhx0416]

Hi!
I train this for indoor scenes. But get depth a little bit weird:
I'm confused is there any mistake I made?

Have you changed the max_depth and min_depth in the options.py. It's better if you can share your setting.

Hi!

I only change the max_depth to 10 in options.py, and the min_depth is 0.1. So what else should I pull the change(like some max_depth settings in training code)?

Hi, hello. Have you figured out the cause of this issue? I've also encountered a similar problem.nianticlabs/monodepth2#483

[lhx0416]

Hi. Have you changed intrinsics?

hello , I met the same issue with him. i changed the intrinsics , and changed the weight\height in option.py,however i the result as follows: RGB image(top) depth map predict from author provided mono_640*192 (middle) trained by myself(bottm)