import matplotlib.pyplot as plt import numpy as np from six import BytesIO from PIL import Image import tensorflow as tf from object_detection.utils import label_map_util from object_detection.utils import visualization_utils as viz_utils from object_detection.utils import ops as utils_op import tarfile import wget import gradio as gr from huggingface_hub import snapshot_download import os #PATH_TO_LABELS = 'data/label_map.pbtxt' PATH_TO_LABELS = 'label_map.pbtxt' category_index = label_map_util.create_category_index_from_labelmap(PATH_TO_LABELS, use_display_name=True) def pil_image_as_numpy_array(pilimg): img_array = tf.keras.utils.img_to_array(pilimg) img_array = np.expand_dims(img_array, axis=0) return img_array def load_image_into_numpy_array(path): image = None image_data = tf.io.gfile.GFile(path, 'rb').read() image = Image.open(BytesIO(image_data)) return pil_image_as_numpy_array(image) def load_model(): download_dir = snapshot_download(REPO_ID) saved_model_dir = os.path.join(download_dir, "saved_model") detection_model = tf.saved_model.load(saved_model_dir) return detection_model #def load_model2(): # wget.download("https://nyp-aicourse.s3-ap-southeast-1.amazonaws.com/pretrained-models/balloon_model.tar.gz") # tarfile.open("balloon_model.tar.gz").extractall() # model_dir = 'saved_model' # detection_model = tf.saved_model.load(str(model_dir)) # return detection_model # samples_folder = 'test_samples # image_path = 'test_samples/sample_balloon.jpeg # def predict(pilimg): image_np = pil_image_as_numpy_array(pilimg) return predict2(image_np) def predict2(image_np): results = detection_model(image_np) # different object detection models have additional results result = {key:value.numpy() for key,value in results.items()} label_id_offset = 0 image_np_with_detections = image_np.copy() viz_utils.visualize_boxes_and_labels_on_image_array( image_np_with_detections[0], result['detection_boxes'][0], (result['detection_classes'][0] + label_id_offset).astype(int), result['detection_scores'][0], category_index, use_normalized_coordinates=True, max_boxes_to_draw=200, min_score_thresh=.60, agnostic_mode=False, line_thickness=2) result_pil_img = tf.keras.utils.array_to_img(image_np_with_detections[0]) return result_pil_img ### def predict_on_video(video_in_filepath, video_out_filepath, detection_model, category_index): video_reader = cv2.VideoCapture(video_in_filepath) frame_h = int(video_reader.get(cv2.CAP_PROP_FRAME_HEIGHT)) frame_w = int(video_reader.get(cv2.CAP_PROP_FRAME_WIDTH)) fps = video_reader.get(cv2.CAP_PROP_FPS) video_writer = cv2.VideoWriter( video_out_filepath, cv2.VideoWriter_fourcc(*'mp4v'), fps, (frame_w, frame_h) ) while True: ret, frame = video_reader.read() if not ret: break # Break the loop if the video is finished processed_frame = predict(frame) processed_frame_np = np.array(processed_frame) video_writer.write(processed_frame_np) # Release camera and close windows video_reader.release() video_writer.release() cv2.destroyAllWindows() cv2.waitKey(1) video_reader.release() video_writer.release() cv2.destroyAllWindows() cv2.waitKey(1) # Function to process a video def process_video(video_path): output_path = "output_video.mp4" # Output path for the processed video predict_on_video(video_path, output_path, detection_model, category_index) return output_path # Specify paths to example images sample_images = [["sample1.jpg"], ["sample2.jpg"], ["sample3.jpg"] ] ### REPO_ID = "gregarific/assignmodel" detection_model = load_model() # pil_image = Image.open(image_path) # image_arr = pil_image_as_numpy_array(pil_image) ### tab1 = gr.Interface(fn=predict, inputs=gr.Image(type="pil"), outputs=gr.Image(type="pil"), examples=[["sample1.jpg"],["sample2.jpg"],["sample3.jpg"]], title="Object Detection (WheelChair & Motorized WheelChair)", description='Model Applied: SSD MobileNet V2 320x320.' ) #gr.Interface(fn=predict, # inputs=gr.Image(type="pil"), # outputs=gr.Image(type="pil") # ).launch(share=True) tab2 = gr.Interface( fn=process_video, inputs=gr.File(label="Upload a Video"), outputs=gr.File(label="Output Analysis"), examples=["Wheelchair Snippet.mp4"], title='Object Detection (WheelChair & Motorized Wheelchair)', description='Model Applied: SSD MobileNet V2 320x320' ) iface = gr.TabbedInterface([tab1, tab2], tab_names = ['Image','Video'], title='WheelChair Type Detection') iface.launch(share=True)