search

noahzn / Lite-Mono

[CVPR2023] Lite-Mono: A Lightweight CNN and Transformer Architecture for Self-Supervised Monocular Depth Estimation
MIT License
527 stars 58 forks source link

about pretainmodel #104

Closed yinyang472 closed 9 months ago

yinyang472 commented 9 months ago

do you have pretainmodel 320*1024 thanks

noahzn commented 9 months ago

Hi, the pre-trained models are resolution agnostic. So, you just need to download the pre-trained model for a specific lite-mono model (small, 8m, etc.)

yinyang472 commented 9 months ago

我有一个问题不太清楚 想和您请教一下 预训练模型输入的大小会影响后续训练时的精度么