Lesson 07: Object Detection with YOLOX

import json
from daisykit.utils import get_asset_file

COCO_CLASSES = [
    "person", "bicycle", "car", "motorcycle", "airplane", "bus", "train",
    "truck", "boat", "traffic light", "fire hydrant", "stop sign",
    "parking meter", "bench", "bird", "cat", "dog", "horse", "sheep", "cow",
    "elephant", "bear", "zebra", "giraffe", "backpack", "umbrella", "handbag",
    "tie", "suitcase", "frisbee", "skis", "snowboard", "sports ball", "kite",
    "baseball bat", "baseball glove", "skateboard", "surfboard", "tennis racket",
    "bottle", "wine glass", "cup", "fork", "knife", "spoon", "bowl", "banana",
    "apple", "sandwich", "orange", "broccoli", "carrot", "hot dog", "pizza",
    "donut", "cake", "chair", "couch", "potted plant", "bed", "dining table",
    "toilet", "tv", "laptop", "mouse", "remote", "keyboard", "cell phone",
    "microwave", "oven", "toaster", "sink", "refrigerator", "book", "clock",
    "vase", "scissors", "teddy bear", "hair drier", "toothbrush",
]

config = {
    "object_detection_model": {
        "model": get_asset_file("models/object_detection/yolox-tiny.param"),
        "weights": get_asset_file("models/object_detection/yolox-tiny.bin"),
        "input_width": 416,
        "input_height": 416,
        "score_threshold": 0.5,   # minimum confidence
        "iou_threshold": 0.8,     # NMS overlap threshold
        "use_gpu": False,
        "class_names": COCO_CLASSES,
    },
}

import cv2
import json
from daisykit.utils import get_asset_file, to_py_type
from daisykit import ObjectDetectorFlow

config = { ... }  # as above

flow = ObjectDetectorFlow(json.dumps(config))
cap = cv2.VideoCapture(0)

while True:
    ret, frame = cap.read()
    if not ret:
        break

    rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)

    objects = flow.Process(rgb)
    flow.DrawResult(rgb, objects)

    display = cv2.cvtColor(rgb, cv2.COLOR_RGB2BGR)
    cv2.imshow("Object Detection (YOLOX)", display)

    if cv2.waitKey(1) & 0xFF == ord("q"):
        break

cap.release()
cv2.destroyAllWindows()

rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
objects = flow.Process(rgb)
objects_py = to_py_type(objects)

for obj in objects_py:
    cls_id   = obj["class_id"]
    cls_name = COCO_CLASSES[cls_id] if cls_id < len(COCO_CLASSES) else "unknown"
    conf     = obj["confidence"]
    x, y, w, h = obj["x"], obj["y"], obj["w"], obj["h"]
    print(f"  {cls_name:15s}  conf={conf:.2f}  bbox=({x},{y},{w},{h})")

WANTED = {"person", "car", "bus", "bicycle"}

objects_py = to_py_type(objects)
people_and_vehicles = [
    obj for obj in objects_py
    if COCO_CLASSES[obj["class_id"]] in WANTED
]

print(f"People and vehicles: {len(people_and_vehicles)}")
for obj in people_and_vehicles:
    print(f"  {COCO_CLASSES[obj['class_id']]}  conf={obj['confidence']:.2f}")

import cv2
import json
import urllib.request
from daisykit.utils import get_asset_file, to_py_type
from daisykit import ObjectDetectorFlow

urllib.request.urlretrieve("https://ultralytics.com/images/bus.jpg", "bus.jpg")

config = { ... }
flow = ObjectDetectorFlow(json.dumps(config))

img = cv2.imread("bus.jpg")
rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)

objects = flow.Process(rgb)
print(f"Detected {len(to_py_type(objects))} objects")

flow.DrawResult(rgb, objects)
cv2.imwrite("detection_result.jpg", cv2.cvtColor(rgb, cv2.COLOR_RGB2BGR))
# Expected: bus.jpg with bounding boxes around ~6 persons, 1 bus, 2 cars

config = {
    "object_detection_model": {
        "model":   "path/to/your_model.param",
        "weights": "path/to/your_model.bin",
        "input_width": 640,
        "input_height": 640,
        "score_threshold": 0.5,
        "iou_threshold": 0.65,
        "use_gpu": False,
        "class_names": ["your_class_1", "your_class_2", "your_class_3"],
    },
}