open-mmlab/mmdetection · error · NotImplementedError

Albu only supports BitMap masks now

Error message

Albu only supports BitMap masks now

What it means

Albu can only pass bitmap (instance) masks to albumentations; PolygonMasks (polygon annotations) are not supported and raise NotImplementedError during _preprocess_results.

Source

Thrown at mmdet/datasets/transforms/transforms.py:1725

    def _preprocess_results(self, results: dict) -> tuple:
        """Pre-processing results to facilitate the use of Albu."""
        if 'bboxes' in results:
            # to list of boxes
            if not isinstance(results['bboxes'], HorizontalBoxes):
                raise NotImplementedError(
                    'Albu only supports horizontal boxes now')
            bboxes = results['bboxes'].numpy()
            results['bboxes'] = [x for x in bboxes]
            # add pseudo-field for filtration
            if self.filter_lost_elements:
                results['idx_mapper'] = np.arange(len(results['bboxes']))

        # TODO: Support mask structure in albu
        ori_masks = None
        if 'masks' in results:
            if isinstance(results['masks'], PolygonMasks):
                raise NotImplementedError(
                    'Albu only supports BitMap masks now')
            ori_masks = results['masks']
            if albumentations.__version__ < '0.5':
                results['masks'] = results['masks'].masks
            else:
                results['masks'] = [mask for mask in results['masks'].masks]

        return results, ori_masks

    def _postprocess_results(
            self,
            results: dict,
            ori_masks: Optional[Union[BitmapMasks,
                                      PolygonMasks]] = None) -> dict:
        """Post-processing Albu output."""
        # albumentations may return np.array or list on different versions
        if 'gt_bboxes_labels' in results and isinstance(
                results['gt_bboxes_labels'], list):

View on GitHub (pinned to cfd5d3a985)

Solutions

1. Set poly2mask=True in LoadAnnotations so masks become bitmap BitMasks
2. Or remove Albu and use native mmdet transforms for mask pipelines
3. Keep Albu only for detection (bbox-only) pipelines

Example fix

# before
pipeline=[dict(type='LoadAnnotations', with_mask=True, poly2mask=False), dict(type='Albu', ...)]
# after
pipeline=[dict(type='LoadAnnotations', with_mask=True, poly2mask=True), dict(type='Albu', ...)]

from mmdet.structures.mask import PolygonMasks
if 'masks' in results:
    assert not isinstance(results['masks'], PolygonMasks), 'Albu needs bitmap masks'

When it happens

Trigger: Using Albu in a pipeline where results['masks'] is a PolygonMasks object (e.g. instance-seg configs loading polygon annotations, typically COCO via LoadAnnotations with poly2mask=False).

Common situations: Adding Albu augmentations to instance segmentation training configs that use polygon masks.

Related errors

• albumentations is not installed
• type must be a str or valid type, but got {type(obj_type)}
• Albu only supports horizontal boxes now
• The `in_channels` of SOLOv2MaskFeatHead and SOLOv2Head shoul
• The annotation file of Open Images Challenge should be a txt

AI-assisted analysis of open-mmlab/mmdetection@cfd5d3a985 (2026-08-27). Data as JSON: /api/errors/487903758e03f7c2. Report an issue: GitHub.