brails.processors.vlm_segmenter.grounded_sam_utils module

class brails.processors.vlm_segmenter.grounded_sam_utils.BoundingBox(xmin: int, ymin: int, xmax: int, ymax: int)

Bases: object

xmax

xmin

property xyxy

ymax

ymin

class brails.processors.vlm_segmenter.grounded_sam_utils.DetectionResult(score: float, label: str, box: brails.processors.vlm_segmenter.grounded_sam_utils.BoundingBox, mask: Optional[<built-in function array>] = None)

Bases: object

box

classmethod from_dict(detection_dict)

label

mask = None

score

brails.processors.vlm_segmenter.grounded_sam_utils.build_models(device='cuda:0')

brails.processors.vlm_segmenter.grounded_sam_utils.detect(image, labels, threshold=0.3, detector_id=None): Use Grounding DINO to detect a set of labels in an image in a zero-shot fashion.

brails.processors.vlm_segmenter.grounded_sam_utils.run_on_one_image(img_source, output_dir, grounding_dino_model, sam_predictor, CLASS_TO_CODE, BOX_THRESHOLD=0.35, TEXT_THRESHOLD=0.25, NMS_THRESHOLD=0.8, visualize=False)

brails.processors.vlm_segmenter.grounded_sam_utils.segment(sam_predictor, image, xyxy)

brails.processors.vlm_segmenter.grounded_sam_utils.show_binary_mask(mask, ax, label_code)

brails.processors.vlm_segmenter.grounded_sam_utils.show_box(box, ax)

brails.processors.vlm_segmenter.grounded_sam_utils.show_points(coords, labels, ax, marker_size=375)

brails.processors.vlm_segmenter.grounded_sam_utils.verify_and_download_models(download_url, filepath)