utils.resize_image got ValueError: height and width must be > 0

[echohenry2006]

When feed large image into the the net , I got "ValueError: height and width must be > 0" in resize_image, anyone know how to solve it?

[gislersoft]

Hey I have same error

/usr/local/lib/python3.6/site-packages/ipykernel_launcher.py:3: DeprecationWarning: `imread` is deprecated!
`imread` is deprecated in SciPy 1.0.0, and will be removed in 1.2.0.
Use ``imageio.imread`` instead.
  This is separate from the ipykernel package so we can avoid doing imports until
Processing 1 images
image                    shape: (480, 640, 3)         min:    0.00000  max:  255.00000
molded_images            shape: (1, 1024, 1024, 3)    min: -123.70000  max:  151.10000
image_metas              shape: (1, 89)               min:    0.00000  max: 1024.00000
/Users/user1/Dev/test/Mask_RCNN/utils.py:67: RuntimeWarning: invalid value encountered in true_divide
  iou = intersection / union
/Users/user1/Dev/test/Mask_RCNN/utils.py:120: RuntimeWarning: invalid value encountered in greater
  remove_ixs = np.where(iou > threshold)[0] + 1
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-7-1c43c85bdf35> in <module>()
      4 
      5 # Run detection
----> 6 results = model.detect([image], verbose=1)
      7 
      8 # Visualize results

~/Dev/test/Mask_RCNN/model.py in detect(self, images, verbose)
   2222             final_rois, final_class_ids, final_scores, final_masks =\
   2223                 self.unmold_detections(detections[i], mrcnn_mask[i],
-> 2224                                        image.shape, windows[i])
   2225             results.append({
   2226                 "rois": final_rois,

~/Dev/test/Mask_RCNN/model.py in unmold_detections(self, detections, mrcnn_mask, image_shape, window)
   2184         for i in range(N):
   2185             # Convert neural network mask to full size mask
-> 2186             full_mask = utils.unmold_mask(masks[i], boxes[i], image_shape)
   2187             full_masks.append(full_mask)
   2188         full_masks = np.stack(full_masks, axis=-1)\

~/Dev/test/Mask_RCNN/utils.py in unmold_mask(mask, bbox, image_shape)
    471     y1, x1, y2, x2 = bbox
    472     mask = scipy.misc.imresize(
--> 473         mask, (y2 - y1, x2 - x1), interp='bilinear').astype(np.float32) / 255.0
    474     mask = np.where(mask >= threshold, 1, 0).astype(np.uint8)
    475 

/usr/local/lib/python3.6/site-packages/numpy/lib/utils.py in newfunc(*args, **kwds)
     99             """`arrayrange` is deprecated, use `arange` instead!"""
    100             warnings.warn(depdoc, DeprecationWarning, stacklevel=2)
--> 101             return func(*args, **kwds)
    102 
    103         newfunc = _set_function_name(newfunc, old_name)

/usr/local/lib/python3.6/site-packages/scipy/misc/pilutil.py in imresize(arr, size, interp, mode)
    562         size = (size[1], size[0])
    563     func = {'nearest': 0, 'lanczos': 1, 'bilinear': 2, 'bicubic': 3, 'cubic': 3}
--> 564     imnew = im.resize(size, resample=func[interp])
    565     return fromimage(imnew)
    566 

/usr/local/lib/python3.6/site-packages/PIL/Image.py in resize(self, size, resample, box)
   1745         self.load()
   1746 
-> 1747         return self._new(self.im.resize(size, resample, box))
   1748 
   1749     def rotate(self, angle, resample=NEAREST, expand=0, center=None,

ValueError: height and width must be > 0

[waleedka]

You might be getting a bounding box with an area of zero, which causes this error when it tries to resize the mask to the given bounding box area. Can you track why your bounding boxes are zero?