Python mask.decode函数代码示例

  def testSingleImageGroundtruthExport(self):
    masks = np.array(
        [[[1, 1,], [1, 1]],
         [[0, 0], [0, 1]],
         [[0, 0], [0, 0]]], dtype=np.uint8)
    boxes = np.array([[0, 0, 1, 1],
                      [0, 0, .5, .5],
                      [.5, .5, 1, 1]], dtype=np.float32)
    coco_boxes = np.array([[0, 0, 1, 1],
                           [0, 0, .5, .5],
                           [.5, .5, .5, .5]], dtype=np.float32)
    classes = np.array([1, 2, 3], dtype=np.int32)
    is_crowd = np.array([0, 1, 0], dtype=np.int32)
    next_annotation_id = 1
    expected_counts = ['04', '31', '4']

    # Tests exporting without passing in is_crowd (for backward compatibility).
    coco_annotations = coco_tools.ExportSingleImageGroundtruthToCoco(
        image_id='first_image',
        category_id_set=set([1, 2, 3]),
        next_annotation_id=next_annotation_id,
        groundtruth_boxes=boxes,
        groundtruth_classes=classes,
        groundtruth_masks=masks)
    for i, annotation in enumerate(coco_annotations):
      self.assertEqual(annotation['segmentation']['counts'],
                       expected_counts[i])
      self.assertTrue(np.all(np.equal(mask.decode(
          annotation['segmentation']), masks[i])))
      self.assertTrue(np.all(np.isclose(annotation['bbox'], coco_boxes[i])))
      self.assertEqual(annotation['image_id'], 'first_image')
      self.assertEqual(annotation['category_id'], classes[i])
      self.assertEqual(annotation['id'], i + next_annotation_id)

    # Tests exporting with is_crowd.
    coco_annotations = coco_tools.ExportSingleImageGroundtruthToCoco(
        image_id='first_image',
        category_id_set=set([1, 2, 3]),
        next_annotation_id=next_annotation_id,
        groundtruth_boxes=boxes,
        groundtruth_classes=classes,
        groundtruth_masks=masks,
        groundtruth_is_crowd=is_crowd)
    for i, annotation in enumerate(coco_annotations):
      self.assertEqual(annotation['segmentation']['counts'],
                       expected_counts[i])
      self.assertTrue(np.all(np.equal(mask.decode(
          annotation['segmentation']), masks[i])))
      self.assertTrue(np.all(np.isclose(annotation['bbox'], coco_boxes[i])))
      self.assertEqual(annotation['image_id'], 'first_image')
      self.assertEqual(annotation['category_id'], classes[i])
      self.assertEqual(annotation['iscrowd'], is_crowd[i])
      self.assertEqual(annotation['id'], i + next_annotation_id)

  def testExportSegmentsToCOCO(self):
    image_ids = ['first', 'second']
    detection_masks = [np.array(
        [[[0, 1, 0, 1], [0, 1, 1, 0], [0, 0, 0, 1], [0, 1, 0, 1]]],
        dtype=np.uint8), np.array(
            [[[0, 1, 0, 1], [0, 1, 1, 0], [0, 0, 0, 1], [0, 1, 0, 1]]],
            dtype=np.uint8)]

    for i, detection_mask in enumerate(detection_masks):
      detection_masks[i] = detection_mask[:, :, :, None]

    detection_scores = [np.array([.8], np.float), np.array([.7], np.float)]
    detection_classes = [np.array([1], np.int32), np.array([1], np.int32)]

    categories = [{'id': 0, 'name': 'person'},
                  {'id': 1, 'name': 'cat'},
                  {'id': 2, 'name': 'dog'}]
    output_path = os.path.join(tf.test.get_temp_dir(), 'segments.json')
    result = coco_tools.ExportSegmentsToCOCO(
        image_ids,
        detection_masks,
        detection_scores,
        detection_classes,
        categories,
        output_path=output_path)
    with tf.gfile.GFile(output_path, 'r') as f:
      written_result = f.read()
      written_result = json.loads(written_result)
      mask_load = mask.decode([written_result[0]['segmentation']])
      self.assertTrue(np.allclose(mask_load, detection_masks[0]))
      self.assertAlmostEqual(result, written_result)

def rle_masks_to_boxes(masks):
    """Computes the bounding box of each mask in a list of RLE encoded masks."""
    if len(masks) == 0:
        return []

    decoded_masks = [
        np.array(mask_util.decode(rle), dtype=np.float32) for rle in masks
    ]

    def get_bounds(flat_mask):
        inds = np.where(flat_mask > 0)[0]
        return inds.min(), inds.max()

    boxes = np.zeros((len(decoded_masks), 4))
    keep = [True] * len(decoded_masks)
    for i, mask in enumerate(decoded_masks):
        if mask.sum() == 0:
            keep[i] = False
            continue
        flat_mask = mask.sum(axis=0)
        x0, x1 = get_bounds(flat_mask)
        flat_mask = mask.sum(axis=1)
        y0, y1 = get_bounds(flat_mask)
        boxes[i, :] = (x0, y0, x1, y1)

    return boxes, np.where(keep)[0]

def evaluate_masks(
    json_dataset,
    all_boxes,
    all_segms,
    output_dir,
    use_salt=True,
    cleanup=False
):
    if cfg.CLUSTER.ON_CLUSTER:
        # On the cluster avoid saving these files in the job directory
        output_dir = '/tmp'
    res_file = os.path.join(
        output_dir, 'segmentations_' + json_dataset.name + '_results')
    if use_salt:
        res_file += '_{}'.format(str(uuid.uuid4()))
    res_file += '.json'

    results_dir = os.path.join(output_dir, 'results')
    if not os.path.exists(results_dir):
        os.mkdir(results_dir)

    os.environ['CITYSCAPES_DATASET'] = DATASETS[json_dataset.name][RAW_DIR]
    os.environ['CITYSCAPES_RESULTS'] = output_dir

    # Load the Cityscapes eval script *after* setting the required env vars,
    # since the script reads their values into global variables (at load time).
    import cityscapesscripts.evaluation.evalInstanceLevelSemanticLabeling \
        as cityscapes_eval

    roidb = json_dataset.get_roidb()
    for i, entry in enumerate(roidb):
        im_name = entry['image']

        basename = os.path.splitext(os.path.basename(im_name))[0]
        txtname = os.path.join(output_dir, basename + 'pred.txt')
        with open(txtname, 'w') as fid_txt:
            if i % 10 == 0:
                logger.info('i: {}: {}'.format(i, basename))
            for j in range(1, len(all_segms)):
                clss = json_dataset.classes[j]
                clss_id = cityscapes_eval.name2label[clss].id
                segms = all_segms[j][i]
                boxes = all_boxes[j][i]
                if segms == []:
                    continue
                masks = mask_util.decode(segms)

                for k in range(boxes.shape[0]):
                    score = boxes[k, -1]
                    mask = masks[:, :, k]
                    pngname = os.path.join(
                        'results',
                        basename + '_' + clss + '_{}.png'.format(k))
                    # write txt
                    fid_txt.write('{} {} {}\n'.format(pngname, clss_id, score))
                    # save mask
                    cv2.imwrite(os.path.join(output_dir, pngname), mask * 255)
    logger.info('Evaluating...')
    cityscapes_eval.main([])
    return None

def polys_to_mask_wrt_box(polygons, box, M):
    """Convert from the COCO polygon segmentation format to a binary mask
    encoded as a 2D array of data type numpy.float32. The polygon segmentation
    is understood to be enclosed in the given box and rasterized to an M x M
    mask. The resulting mask is therefore of shape (M, M).
    """
    w = box[2] - box[0]
    h = box[3] - box[1]

    w = np.maximum(w, 1)
    h = np.maximum(h, 1)

    polygons_norm = []
    for poly in polygons:
        p = np.array(poly, dtype=np.float32)
        p[0::2] = (p[0::2] - box[0]) * M / w
        p[1::2] = (p[1::2] - box[1]) * M / h
        polygons_norm.append(p)

    rle = mask_util.frPyObjects(polygons_norm, M, M)
    mask = np.array(mask_util.decode(rle), dtype=np.float32)
    # Flatten in case polygons was a list
    mask = np.sum(mask, axis=2)
    mask = np.array(mask > 0, dtype=np.float32)
    return mask

def crop_mask(boxes,segmentations,flipped, imsize):
    assert (boxes.shape[0]==len(segmentations))
    psegmentations=[]
    for i in xrange(len(segmentations)):
        gts=segmentations[i]
        box=boxes[i,:]
        if type(gts) == list and gts:
            assert (type(gts[0]) != dict)
            prle= mask.frPyObjects(gts,imsize[1],imsize[0])
        elif type(gts) == dict and type(gts['counts']) == list:
            prle= mask.frPyObjects([gts],imsize[1],imsize[0])
        elif type(gts) == dict and \
                     type(gts['counts'] == unicode or type(gts['counts']) == str):
            prle = [gts]
        else:
            print '{} box has no segmentation'.format(i)
            psegmentations.append([])
            continue
        if len(prle)==1:
            prle=prle[0]
        else:
            prle= mask.merge(prle)
        pmask=mask.decode([prle])
        if flipped:
            pmask=pmask[:,::-1,:]
        pmask=np.copy(pmask[box[1]:box[3],box[0]:box[2],:],order='F')
        psegmentations.append(mask.encode(pmask))
    return psegmentations

    def load_dataset(self):
        dataset  = self.cfg.dataset
        dataset_phase = self.cfg.dataset_phase
        dataset_ann = self.cfg.dataset_ann

        # initialize COCO api
        annFile = '%s/annotations/%s_%s.json'%(dataset,dataset_ann,dataset_phase)
        self.coco = COCO(annFile)

        imgIds = self.coco.getImgIds()

        data = []

        # loop through each image
        for imgId in imgIds:
            item = DataItem()

            img = self.coco.loadImgs(imgId)[0]
            item.im_path = "%s/images/%s/%s"%(dataset, dataset_phase, img["file_name"])
            item.im_size = [3, img["height"], img["width"]]
            item.coco_id = imgId
            annIds = self.coco.getAnnIds(imgIds=img['id'], iscrowd=False)
            anns = self.coco.loadAnns(annIds)

            all_person_keypoints = []
            masked_persons_RLE = []
            visible_persons_RLE = []
            all_visibilities = []

            # Consider only images with people
            has_people = len(anns) > 0
            if not has_people and self.cfg.coco_only_images_with_people:
                continue

            for ann in anns: # loop through each person
                person_keypoints = []
                visibilities = []
                if ann["num_keypoints"] != 0:
                    for i in range(self.cfg.num_joints):
                        x_coord = ann["keypoints"][3 * i]
                        y_coord = ann["keypoints"][3 * i + 1]
                        visibility = ann["keypoints"][3 * i + 2]
                        visibilities.append(visibility)
                        if visibility != 0: # i.e. if labeled
                            person_keypoints.append([i, x_coord, y_coord])
                    all_person_keypoints.append(np.array(person_keypoints))
                    visible_persons_RLE.append(maskUtils.decode(self.coco.annToRLE(ann)))
                    all_visibilities.append(visibilities)
                if ann["num_keypoints"] == 0:
                    masked_persons_RLE.append(self.coco.annToRLE(ann))

            item.joints = all_person_keypoints
            item.im_neg_mask = maskUtils.merge(masked_persons_RLE)
            if self.cfg.use_gt_segm:
                item.gt_segm = np.moveaxis(np.array(visible_persons_RLE), 0, -1)
                item.visibilities = all_visibilities
            data.append(item)

        self.has_gt = self.cfg.dataset is not "image_info"
        return data

 def compute_scmap_weights(self, scmap_shape, joint_id, data_item):
     size = scmap_shape[0:2]
     scmask = np.ones(size)
     m = maskUtils.decode(data_item.im_neg_mask)
     if m.size:
         scmask = 1.0 - imresize(m, size)
     scmask = np.stack([scmask] * self.cfg.num_joints, axis=-1)
     return scmask

 def annToMask(self, ann, height, width):
     """
     Convert annotation which can be polygons, uncompressed RLE, or RLE to binary mask.
     :return: binary mask (numpy 2D array)
     """
     rle = self.annToRLE(ann, height, width)
     m = maskUtils.decode(rle)
     return m

 def _flip_rle(rle, height, width):
     if 'counts' in rle and type(rle['counts']) == list:
         # Magic RLE format handling painfully discovered by looking at the
         # COCO API showAnns function.
         rle = mask_util.frPyObjects([rle], height, width)
     mask = mask_util.decode(rle)
     mask = mask[:, ::-1, :]
     rle = mask_util.encode(np.array(mask, order='F', dtype=np.uint8))
     return rle

    def draw_objdb_masks(self, output_dir, objdb=None):
        if objdb == None:
            objdb = self.objdb

        mask_dir = osp.join(output_dir, '{}_objdb_masks'.format(self._image_set))
        img_dir  = osp.join(output_dir, '{}_objdb_imgs'.format(self._image_set))

        ds_utils.maybe_create(output_dir)
        ds_utils.maybe_create(mask_dir)
        ds_utils.maybe_create(img_dir)


        for i in xrange(len(objdb)):
            obj     = objdb[i]

            im_path = obj['image']
            ann_id  = obj['obj_id']
            poly    = obj['poly']
            bb      = obj['box'].astype(np.int16)
            cls     = obj['cls']
            width   = obj['width']
            height  = obj['height']

            img = cv2.imread(im_path, cv2.IMREAD_COLOR)
            msk = np.amax(COCOmask.decode(poly), axis=2)

            # binarize the mask
            msk = msk * 255
            retVal, msk = cv2.threshold(msk, 127, 255, cv2.THRESH_BINARY)
            msk = msk.astype(np.uint8)
            # msk = ds_utils.dilate_mask(msk, 9)

            # img = (1 - 0.5/255 * msk.reshape((height, width, 1))) * img + \
            #       0.5/255 * msk.reshape((height, width, 1)) * \
            #       np.random.random((1, 3)) * 255

            # cv2.rectangle(img, (bb[0], bb[1]), (bb[2], bb[3]), \
            #             (0, 255, 0), 2)
            #
            # fontScale = 0.0009 * math.sqrt(float(width*width + height*height))
            #
            #
            # cv2.putText(img, '{:}'.format(self.classes[cls]), \
            #             (bb[0], bb[1] - 2), \
            #             cv2.FONT_HERSHEY_SIMPLEX, \
            #             fontScale, (0, 0, 255), 1)

            im_name, im_ext = osp.splitext(osp.basename(im_path))

            output_path = osp.join(mask_dir, im_name+'_'+str(ann_id).zfill(12)+im_ext)
            # output_path = osp.join(mask_dir, im_name+im_ext)
            cv2.imwrite(output_path, msk)

            output_path = osp.join(img_dir,  im_name+'_'+str(ann_id).zfill(12)+im_ext)
            # output_path = osp.join(img_dir,  im_name+im_ext)
            cv2.imwrite(output_path, img)
            print i

def get_mask(idx):
    ann_ids = coco.getAnnIds(imgIds=img_ids[idx])
    anns = coco.loadAnns(ann_ids)
    img = coco.loadImgs(img_ids[idx])[0]
    m = np.zeros((img['height'], img['width']))
    for j in anns:
        if j['iscrowd']:
            rle = mask.frPyObjects(j['segmentation'], img['height'], img['width'])
            m += mask.decode(rle)
    return m < 0.5

 def convert(self, mode):
     width, height = self.size
     if mode == "mask":
         rles = mask_utils.frPyObjects(
             [p.numpy() for p in self.polygons], height, width
         )
         rle = mask_utils.merge(rles)
         mask = mask_utils.decode(rle)
         mask = torch.from_numpy(mask)
         # TODO add squeeze?
         return mask

    def draw_roidb_masks(self, output_dir, roidb=None):

        mask_dir = osp.join(output_dir, '{}_roidb_masks'.format(self._image_set))
        img_dir  = osp.join(output_dir, '{}_roidb_imgs'.format(self._image_set))

        ds_utils.maybe_create(output_dir)
        ds_utils.maybe_create(mask_dir)
        ds_utils.maybe_create(img_dir)

        if roidb == None:
            roidb = self.roidb

        for i in xrange(len(roidb)):
            rois    = roidb[i]
            im_path = rois['image']
            clses   = rois['clses']
            boxes   = rois['boxes']
            rles    = rois['polys']
            width   = rois['width']
            height  = rois['height']

            img = cv2.imread(im_path, cv2.IMREAD_COLOR)
            msk = np.zeros((height, width), dtype=np.uint8)

            for j in xrange(len(rles)):
                rle = rles[j]
                bb  = boxes[j,:].astype(np.int)
                cls = clses[j]

                tmp = np.amax(COCOmask.decode(rle), axis=2) * 255
                retVal, tmp = cv2.threshold(tmp, 127, 255, cv2.THRESH_BINARY)
                tmp = tmp.astype(np.uint8)
                tmp = ds_utils.dilate_mask(tmp, 9)
                msk = np.maximum(msk, tmp)

                # fontScale = 0.0009 * math.sqrt(float(width*width + height*height))
                # cv2.rectangle(img, (bb[0], bb[1]), (bb[2], bb[3]), \
                #             (0, 255, 0), 2)
                # cv2.putText(img, '{:}'.format(self.classes[cls]), \
                #             (bb[0], bb[1] - 2), \
                #             cv2.FONT_HERSHEY_SIMPLEX, \
                #             fontScale, (0, 0, 255), 1)

            # img = (1 - 0.5/255 * msk.reshape((height, width, 1))) * img + \
            #       0.5/255 * msk.reshape((height, width, 1)) * \
            #       np.random.random((1, 3)) * 255


            output_path = osp.join(mask_dir, osp.basename(im_path))
            cv2.imwrite(output_path, msk)
            output_path = osp.join(img_dir,  osp.basename(im_path))
            cv2.imwrite(output_path, img)

            print i

def polys_to_mask(polygons, height, width):
    """Convert from the COCO polygon segmentation format to a binary mask
    encoded as a 2D array of data type numpy.float32. The polygon segmentation
    is understood to be enclosed inside a height x width image. The resulting
    mask is therefore of shape (height, width).
    """
    rle = mask_util.frPyObjects(polygons, height, width)
    mask = np.array(mask_util.decode(rle), dtype=np.float32)
    # Flatten in case polygons was a list
    mask = np.sum(mask, axis=2)
    mask = np.array(mask > 0, dtype=np.float32)
    return mask