noahzn / Lite-Mono

[CVPR2023] Lite-Mono: A Lightweight CNN and Transformer Architecture for Self-Supervised Monocular Depth Estimation
MIT License
540 stars 61 forks source link

如何让模型适配鱼眼图片呢? #105

Closed Linkersem closed 10 months ago

Linkersem commented 10 months ago

您好,我想请问一下我能否直接用鱼眼图片训练来使模型能够对鱼眼图片进行深度预测呢?

noahzn commented 10 months ago

你好,我没有试过使用鱼眼图片,但是我觉得整体方法是可行。只是可能鱼眼图像有较大的图像畸变需要处理,而且应该需要对代码中的图像重建部分进行调整以使鱼眼图像能够顺利重建。

Linkersem commented 10 months ago

谢谢您的回复,能麻烦您展开讲一下”对代码中的图像重建部分进行调整以使鱼眼图像能够顺利重建。“这一部分的思想或者逻辑吗?这一部分我没有get到

noahzn commented 10 months ago

Lite-Mono是基于Monodepth2的,将深度估计转换为图像重建任务并优化loss。你要自己测试一下那部分代码能不能直接用在你的鱼眼图像上。

noahzn commented 10 months ago

I am now closing this issue due to a lack of response.