from transformers import DetrImageProcessor, DetrForObjectDetection
import torch
from PIL import Image
import gradio as gr

def detect_objects(image):
    # Load the pre-trained DETR model
    processor = DetrImageProcessor.from_pretrained("facebook/detr-resnet-101")
    model = DetrForObjectDetection.from_pretrained("facebook/detr-resnet-101")

    inputs = processor(images=image, return_tensors="pt")
    outputs = model(**inputs)

    # convert outputs (bounding boxes and class logits) to COCO API
    # let's only keep detections with score > 0.9
    target_sizes = torch.tensor([image.size[::-1]])
    results = processor.post_process_object_detection(outputs, target_sizes=target_sizes, threshold=0.9)[0]

    res = []
    for label in results["labels"]:
        res.append(model.config.id2label[label.item()])
        
    return ','.join(res)

def upload_image(file):
    image = Image.open(file.name)
    image_with_boxes = detect_objects(image)
    return image_with_boxes

iface = gr.Interface(
    fn=upload_image,
    inputs="file",
    outputs="text",
    title="Object Detection",
    description="Upload an image and detect objects using DETR model.",
    flagging_mode="never"
)

iface.launch()