roboflow/supervision · error · ValueError

EasyOCR results produced with detail=0 do not include boundi

Error message

EasyOCR results produced with detail=0 do not include bounding boxes. Call reader.readtext(..., detail=1) instead.

What it means

Detections.from_easyocr expects EasyOCR's detail=1 output format, where each item is (bbox_points, text, confidence). When EasyOCR is called with detail=0 it returns plain strings with no bounding boxes, which supervision detects by checking the first element's type and rejects.

Source

Thrown at src/supervision/detection/core.py:2174

        Returns:
            A new Detections object.

        Example:
            ```python
            import supervision as sv
            import easyocr

            reader = easyocr.Reader(['en'])
            results = reader.readtext("<SOURCE_IMAGE_PATH>")
            detections = sv.Detections.from_easyocr(results)
            detected_text = detections["class_name"]
            ```
        """
        if len(easyocr_results) == 0:
            return cls.empty()

        if isinstance(easyocr_results[0], str):
            raise ValueError(
                "EasyOCR results produced with detail=0 do not include bounding "
                "boxes. Call reader.readtext(..., detail=1) instead."
            )

        bbox = np.array([result[0] for result in easyocr_results], dtype=np.float32)
        if bbox.ndim != 3 or bbox.shape[1:] != (4, 2):
            raise ValueError(
                "EasyOCR results must contain four corner points per detection."
            )
        xyxy = np.hstack((np.min(bbox, axis=1), np.max(bbox, axis=1)))
        confidence = np.array(
            [
                result[2] if len(result) > 2 and result[2] else 0
                for result in easyocr_results
            ]
        )
        ocr_text = np.array([result[1] for result in easyocr_results])

View on GitHub (pinned to 7f254d9784)

Solutions

1. Call reader.readtext(image, detail=1) (detail=1 is the default — remove any explicit detail=0).
2. Pass the resulting list of (quad_points, text, score) tuples straight to from_easyocr.

Example fix

# before
results = reader.readtext('image.jpg', detail=0)
detections = sv.Detections.from_easyocr(results)

# after
results = reader.readtext('image.jpg', detail=1)
detections = sv.Detections.from_easyocr(results)

def easyocr_ready(results: list) -> bool:
    return len(results) == 0 or not isinstance(results[0], str)

results = reader.readtext(image, detail=1)  # detail=1 is required
assert easyocr_ready(results)
detections = sv.Detections.from_easyocr(results)

def is_easyocr_detail_1(results: list) -> bool:
    return all(
        isinstance(r, (list, tuple)) and len(r) >= 2 and isinstance(r[0], (list, tuple))
        for r in results
    )

try:
    detections = sv.Detections.from_easyocr(results)
except ValueError as e:
    if 'detail=0' in str(e):
        results = reader.readtext(image, detail=1)
        detections = sv.Detections.from_easyocr(results)
    else:
        raise

When it happens

Trigger: Calling reader.readtext(image, detail=0) — or a reader configured with detail=0 — and passing the string list to Detections.from_easyocr. The isinstance(easyocr_results[0], str) check trips and raises.

Common situations: Developers previously used EasyOCR just for text extraction (detail=0 is common in text-only pipelines) and reuse the same call when adding detection visualization; copy-pasted snippets from text-summarization code.

Related errors

• EasyOCR results must contain four corner points per detectio

AI-assisted analysis of roboflow/supervision@7f254d9784 (2026-08-15). Data as JSON: /api/errors/8c450256b4059664. Report an issue: GitHub.