Comparison of model learning and thesis results

[taewookum]

I have recently implemented the depth estimation model described in your paper using the code provided in this repository. However, I have encountered a significant discrepancy between the performance metrics I achieved and the results reported in your paper.

I've completed my learning normally as described, is there a problem?

Command : python train.py --config configs/kitti_raw.yaml

[pandaswfas]

I also interesting in this paper, can you tell me what is the accuracy of your reproduction?

[taewookum]

I also interesting in this paper, can you tell me what is the accuracy of your reproduction?

Scaling ratios | med: 34.1213 | std: 0.0866

abs_rel | sq_rel | rmse | rmse_log | a1 | a2 | a3 |
& 0.1158 & 0.8095 & 4.6254 & 0.1901 & 0.8708 & 0.9610 & 0.9828 \

This is the result of learning the model I was provided with.

[fengyi233]

Thanks for the reminder! Some bugs have been found and I will fix them as soon as possible.

[wangjiyuan9]

Have you ever fixed the bug?