app.py

import gradio as gr
from transformers import DPTFeatureExtractor, DPTForDepthEstimation
import torch
import numpy as np
from PIL import Image
import open3d as o3d
from pathlib import Path
import os

feature_extractor = DPTFeatureExtractor.from_pretrained("Intel/dpt-large")
model = DPTForDepthEstimation.from_pretrained("Intel/dpt-large")


def process_image(image_path):
    image_path = Path(image_path)
    image = Image.open(image_path)

    # prepare image for the model
    encoding = feature_extractor(image, return_tensors="pt")

    # forward pass
    with torch.no_grad():
        outputs = model(**encoding)
        predicted_depth = outputs.predicted_depth

    # interpolate to original size
    prediction = torch.nn.functional.interpolate(
        predicted_depth.unsqueeze(1),
        size=image.size[::-1],
        mode="bicubic",
        align_corners=False,
    ).squeeze()
    output = prediction.cpu().numpy()
    depth_image = (output * 255 / np.max(output)).astype("uint8")
    img = Image.fromarray(depth_image)
    return [img]



title = "Demo: zero-shot depth estimation with DPT + 3D Point Cloud"
description = "This demo is a variation from the original <a href='https://huggingface.co/spaces/nielsr/dpt-depth-estimation' target='_blank'>DPT Demo</a>. It uses the DPT model to predict the depth of an image and then uses 3D Point Cloud to create a 3D object."
examples = [["examples/" + img] for img in os.listdir("examples/")]

iface = gr.Interface(
    fn=process_image,
    inputs=[gr.inputs.Image(type="filepath", label="Input Image")],
    outputs=[gr.outputs.Image(type="pil", label="Predicted Depth")],
    title=title,
    description=description,
    examples=examples,
    allow_flagging="never",
    cache_examples=False,
)
iface.launch(debug=True, show_api=False)