keras-team/keras · error · ValueError

Expected `bounding_boxes['boxes']` to have rank 2 or 3, with

Error message

Expected `bounding_boxes['boxes']` to have rank 2 or 3, with shape (num_boxes, 4) or (batch_size, num_boxes, 4). Received: bounding_boxes['boxes'].shape={boxes_shape}

What it means

validate_bounding_boxes only accepts dense 'boxes' of rank 2 ((num_boxes, 4), unbatched) or rank 3 ((batch, num_boxes, 4), batched). Rank-1, rank-4, or higher boxes tensors cannot be interpreted as a set of xyxy boxes and are rejected.

Source

Thrown at keras/src/layers/preprocessing/image_preprocessing/bounding_boxes/validation.py:176

            if len(labels_shape) not in {1, 2}:
                raise ValueError(
                    "Found "
                    f"bounding_boxes['boxes'].shape={boxes_shape} "
                    "and expected bounding_boxes['labels'] to have "
                    "rank 1 or 2, but received: "
                    f"bounding_boxes['labels'].shape={labels_shape} "
                )
        elif len(boxes_shape) == 3:
            if len(labels_shape) not in {2, 3}:
                raise ValueError(
                    "Found "
                    f"bounding_boxes['boxes'].shape={boxes_shape} "
                    "and expected bounding_boxes['labels'] to have "
                    "rank 2 or 3, but received: "
                    f"bounding_boxes['labels'].shape={labels_shape} "
                )
        else:
            raise ValueError(
                "Expected `bounding_boxes['boxes']` "
                "to have rank 2 or 3, with shape "
                "(num_boxes, 4) or (batch_size, num_boxes, 4). "
                "Received: "
                f"bounding_boxes['boxes'].shape={boxes_shape}"
            )

View on GitHub (pinned to 7a34a03db6)

Solutions

1. Squeeze stray leading dims: boxes = keras.ops.squeeze(boxes, axis=0) until rank is 2 or 3
2. For a single image ensure shape (num_boxes, 4); for a batch (batch, num_boxes, 4)
3. Print boxes.shape right before the call to spot the extra axis

Example fix

# before
boxes = np.array([all_boxes])  # shape (1, batch, boxes, 4) -> rank 4
# after
boxes = np.array(all_boxes)    # shape (batch, boxes, 4)

assert len(bbs['boxes'].shape) in (2, 3), bbs['boxes'].shape

When it happens

Trigger: densify_bounding_boxes with boxes of rank 1 (flat 8-vector), rank 4 (extra leading dims, e.g. (1, batch, boxes, 4)).

Common situations: Residual batch dim of 1 left after expand_dims; wrapping once more in np.array([x]) making rank 4; feeding keypoint arrays of the wrong rank.

Related errors

• Found bounding_boxes['boxes'].shape={boxes_shape} and expect
• If `bounding_boxes['boxes']` is a list, then `bounding_boxes
• Found bounding_boxes['boxes'].shape={boxes_shape} and expect
• Expected as input a list/tuple of 2 tensors. Received input_
• `height` and `width` must be set if `format='xyxy'`.

AI-assisted analysis of keras-team/keras@7a34a03db6 (2026-08-25). Data as JSON: /api/errors/9f23f506da0a44a8. Report an issue: GitHub.