labels or label isn't an existing parameter for box_annotate()

[Jarradmorden]

Search before asking

• I have searched the Supervision issues and found no similar bug report.

Bug

labels or label isn't an existing parameter for box_annotate(), what do we replace this with? is it needed I just removed it?

Notebook = https://github.com/roboflow/notebooks/blob/main/notebooks/how-to-auto-train-yolov8-model-with-autodistill.ipynb

box_annotator = sv.BoxAnnotator()
image = cv2.imread(f"C:/Users/jmorde02/DATA/2024-10-18_15-13-59\sample\left_3070_2024-10-18 15-17-30.481540.png")

mask_annotator = sv.MaskAnnotator()
results = base_model.predict(image)

annotated_image = mask_annotator.annotate(
image.copy(), detections=results
)

images = []
for image_name in image_names:
image = dataset.images[image_name]
annotations= dataset.annotations[image_name]
labels = [
dataset.classes[class_id]
for class_id
in annotations.class_id]
annotates_image = mask_annotator.annotate(
scene=image.copy(),
detections=annotations)
annotates_image = box_annotator.annotate(
scene=annotates_image,
detections=annotations
labels=labels)
images.append(annotates_image)

As you can see here the options

(method) def annotate(
scene: ImageType@annotate,
detections: Detections,
custom_color_lookup: ndarray | None = None
) -> ImageType@annotate

Environment

supervision==0.24.0
Windows 11
Python3.11
Cuda 11.8

Minimal Reproducible Example

No response

Additional

No response

Are you willing to submit a PR?

• Yes I'd like to help by submitting a PR!

[LinasKo]

Hi @Jarradmorden 👋

The supervision version there is indeed very old. Here's how you'd use the annotators right now:

mask_annotator = sv.MaskAnnotator()
box_annotator = sv.BoxAnnotator()
label_annotator = sv.LabelAnnotator()

images = []
for image_name, image, annotations in dataset:
    annotated_image = image.copy()
    annotated_image = mask_annotator.annotate(
        scene=annotated_image,
        detections=annotations)
    annotated_image = box_annotator.annotate(
        scene=annotated_image,
        detections=annotations)
    annotated_image = label_annotator.annotate(
        scene=annotated_image,
        detections=annotations,
    )
    images.append(annotated_image)

The LabelAnnotator.annotate() does accept a labels parameter, but you probably won't need it.

[onuralpszr]

Converting to Q&A since this is not a bug

[onuralpszr]

Okay then sure do as you said 👍 but make sure test as well

[Jarradmorden]

I did test but because of low sample I couldn't tell if it was doing it at the time as it was only doing the car on front of me ! Hah thanks very much !

[onuralpszr]

@Jarradmorden İf side note original question answer solve your problem please mark as answer 👍 That would be nice

[Jarradmorden]

Awesome will do thanks

[LinasKo]

To add more ideas into the mix, check out the dataset used for this one: https://github.com/roboflow/notebooks/blob/main/notebooks/train-yolov8-instance-segmentation-on-custom-dataset.ipynb