depth_detection.py

import torch
from transformers import GLPNImageProcessor, GLPNForDepthEstimation
from image_resize import resize_img

def depth_detection(image, pad=16):

    feature_extractor = GLPNImageProcessor.from_pretrained("vinvino02/glpn-nyu")
    model = GLPNForDepthEstimation.from_pretrained("vinvino02/glpn-nyu")
    new_img = resize_img(image)
    inputs = feature_extractor(images=new_img, return_tensors="pt")

    with torch.no_grad():
        outputs = model(**inputs)
        predicted_depth = outputs.predicted_depth

    output = predicted_depth.squeeze().cpu().numpy() * 1000.0
    output = output[pad:-pad, pad:-pad]
    new_image = new_img.crop((pad, pad, new_img.width - pad, new_img.height - pad))
    return new_image,output