ami-detector-boxes
All format examples · Machine-readable schema · Convert outputs to COCO
One image: integer pixel corners. The wrapper discards scores and classes. Supply image identity and dimensions separately.
Example
Illustrative values, not predictions from a model. Unless stated otherwise, the image is 200 pixels wide and 100 pixels high.
Readable projection: arrays and objects below illustrate the contents. The author returns Python objects/tensors or files, rather than this JSON document.
[
[10, 20, 50, 60]
]Convert to a standard format
Use Converter.convert with source="ami-detector-boxes" and target="coco". Your application supplies the predictions and any required image context or producer settings. The adapter validates whether the output contains the information the target needs.
General conversion example and required context
Inspect the intermediate representation for this example
This developer example prints ISIR, the internal representation used between adapters. Run from the repository checkout; no model download is needed.
python3 examples/formats/to_isir.py ami-detector-boxesHow to interpret it
Scope: One image after author thresholding; post_process_batch adds an outer image list. Excludes raw torchvision tensors, database records and downstream API detections/classifications.
Coordinates: [x1, y1, x2, y2], original-image pixels, origin top-left. NumPy astype(int) truncates the retained coordinates toward zero.
Confidence: The author filters scores strictly greater than 0.80 (2023 ResNet50) or 0.50 (MobileNetV3), then discards scores and labels. Neither confidence nor class is present in this output.
Empty: No retained detections produces [].
Evidence
Source inspection at the pinned revisions; no full inference observation claimed.