app.py

import gradio as gr
from transformers import AutoImageProcessor, ResNetForImageClassification, YolosFeatureExtractor, YolosForObjectDetection
import torch




def detect(image1, image2):
    ### Image 1, the object ### 
    processor = AutoImageProcessor.from_pretrained("microsoft/resnet-50")
    model = ResNetForImageClassification.from_pretrained("microsoft/resnet-50")
    
    inputs = processor(image1, return_tensors="pt")
    
    with torch.no_grad():
        logits = model(**inputs).logits
    
    # model predicts one of the 1000 ImageNet classes
    predicted_label = logits.argmax(-1).item()
    object_label = model.config.id2label[predicted_label]
    
    ### Image 2, object detections ###
    from PIL import Image
    import requests
    
    feature_extractor = YolosFeatureExtractor.from_pretrained('hustvl/yolos-small')
    model = YolosForObjectDetection.from_pretrained('hustvl/yolos-small')
    
    inputs = feature_extractor(images=image2, return_tensors="pt")
    outputs = model(**inputs)
    
    # model predicts bounding boxes and corresponding COCO classes
    logits = outputs.logits
    bboxes = outputs.pred_boxes

    return object_label, bboxes




demo = gr.Interface(
    fn=detect,
    inputs=[gr.inputs.Image(label="Object to detect"), gr.inputs.Image(label="Image to detect object in")],
    outputs=["text", "text"],
    title="Object Counts in Image"
)

demo.launch()