app.py

from io import BytesIO
from tempfile import NamedTemporaryFile

from fastapi import FastAPI, Response, status, UploadFile
from torchvision.io import read_image
from torchvision.models.detection import (FasterRCNN_ResNet50_FPN_V2_Weights,
                                          fasterrcnn_resnet50_fpn_v2)
from torchvision.transforms.v2.functional import to_pil_image
from torchvision.utils import draw_bounding_boxes
from PIL import Image

app = FastAPI(docs_url='/', title='Test PyTorch COCO Object Detection')

# Step 1: Initialize model with the best available weights
weights = FasterRCNN_ResNet50_FPN_V2_Weights.DEFAULT
model = fasterrcnn_resnet50_fpn_v2(weights=weights, box_score_thresh=0.9)
model.eval()

# Step 2: Initialize the inference transforms
preprocess = weights.transforms()


@app.get('/healthcheck')
async def healthcheck():
    return Response(status_code=status.HTTP_200_OK)


@app.post('/detectObjects')
async def infer(image: UploadFile):
    with NamedTemporaryFile() as f:
        f.write(image.file.read())
        f.seek(0)
        img = read_image(f.name)

    batch = [preprocess(img)]

    prediction = model(batch)[0]
    labels = [weights.meta["categories"][i] for i in prediction["labels"]]
    box = draw_bounding_boxes(img, boxes=prediction["boxes"],
                              labels=labels,
                              colors="red",
                              width=4, font_size=30)
    im = to_pil_image(box.detach())
    with BytesIO() as bio:
        im.save(bio, format='PNG')
        return Response(content=bio.getvalue(), media_type='image/png')