HybridDepth

HybridDepth copied to clipboard

Correct me if I am wrong: 1. During inference, we need to prepare focal stack of 5 images for each scene. 2. During training we need only one image with...

SuteraBlu

I attempted to use a single image with the trained models, but it was unsuccessful. Is there a way to generate a depth model using only one image?

Seifwael2

from utils.io import prepare_input_image data_dir = 'focal stack images directory' ####Load the focal stack images focal_stack, rgb_img, focus_dist = prepare_input_image(data_dir) 使用prepare_input_image进行预处理时，如 focus_dist = np.append(focus_dist, float(image_files[0].split("_")[-1].removesuffix(".png"))) 这里把第一这张图片名作为焦距是为什么呢，另外可以解释下预处理流程吗 （萌新请教 谢谢大佬）

LeoRainly

Thank you for sharing your work on HYBRIDDEPTH. It’s impressive and fits perfectly with my AR mobile app project in Unity. Currently, I use Unity Sentis with a [relative depth...

ocularbioengarhku