ValueError：所有边界框的高度和宽度都应为正

Question

如有任何帮助，我们将不胜感激。 我知道为什么会发生错误，这是因为xmin==xmax和ymin==ymax不应该是。然而，我似乎不知道这是怎么发生的。下面是如何使用pytorch数据集类加载自定义数据集

`class CustomDataset(torch.utils.data.Dataset):
    def __init__(self, root, transforms=None):
        self.root = root
        self.transforms = transforms
        # load all image files, sorting them to
        # ensure that they are aligned
        self.imgs = list(sorted(os.listdir(os.path.join(root, "seg_image_use"))))
        self.masks = list(sorted(os.listdir(os.path.join(root, "seg_mask_use"))))

    def __getitem__(self, idx):
        # load one image and mask using idx
        img_path = os.path.join(self.root, "seg_image_use", self.imgs[idx])
        mask_path = os.path.join(self.root, "seg_mask_use", self.masks[idx])
        img = Image.open(img_path).convert("RGB")
        # note that we haven't converted the mask to RGB,
        # because each color corresponds to a different instance
        # with 0 being background
        mask = Image.open(mask_path)

        mask = np.asarray(mask)
        # instances are encoded as different colors
        obj_ids = np.unique(mask)[1:] # first id is the background, so remove it   
        masks = mask == obj_ids[:, None, None]  # split the color-encoded mask into a set of binary masks
        # get bounding box coordinates for each mask
        num_objs = len(obj_ids)
        boxes = []
        for i in range(num_objs):
            pos = np.where(masks[i])
            xmin = np.min(pos[1])
            xmax = np.max(pos[1])
            ymin = np.min(pos[0])
            ymax = np.max(pos[0])
            boxes.append([xmin, ymin, xmax, ymax])

       # convert everything into torch.Tensor
        boxes = torch.as_tensor(boxes, dtype=torch.float32)      
        area = (boxes[:, 3] - boxes[:, 1]) * (boxes[:, 2] - boxes[:, 0])

        target = {}
        target["boxes"] = boxes
        target["labels"] = torch.as_tensor(obj_ids, dtype=torch.int64) - 1 # corrected by Rawi
        target["masks"] = torch.as_tensor(masks, dtype=torch.uint8) #uint8
        target["image_id"] = torch.tensor([idx]) 
        target["area"] = area
        target["iscrowd"] = torch.zeros((num_objs,), dtype=torch.int64) # suppose all instances are not crowd
        
        if self.transforms is not None:
            img, target = self.transforms(img, target)

        return img, target

    def __len__(self):
        return len(self.imgs)
`
And then when I call it to see the dataset, I get the first index of each dataset showing this; take note of the first index in the 'boxes' tensor. (italic)

`dataset_sample = CustomDataset('C:/Users/LENOVO/Desktop/clothme/Train')
img, target = dataset_sample[2]
print(target)

result: {'boxes': tensor(*[[  0.,   0., 286., 403.]*,
         [ 30., 240.,  52., 241.],
         [ 25., 183.,  31., 204.],
         [ 26., 224.,  34., 240.],
         [ 30., 169.,  88., 181.],
         [ 32., 239.,  85., 251.],
`

这是我尝试训练模型时遇到的错误

`---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-81-c798930961c1> in <module>
      4 for epoch in range(num_epochs):
      5     # train for one epoch, printing every 10 iterations
----> 6     train_one_epoch(model, optimizer, data_loader, device, epoch, print_freq=10)
      7     # update the learning rate
      8     lr_scheduler.step()

~\measurement_model_dev\engine.py in train_one_epoch(model, optimizer, data_loader, device, epoch, print_freq)
     28         targets = [{k: v.to(device) for k, v in t.items()} for t in targets]
     29 
---> 30         loss_dict = model(images, targets)
     31 
     32         losses = sum(loss for loss in loss_dict.values())

~\anaconda3\envs\measurement_py37\lib\site-packages\torch\nn\modules\module.py in _call_impl(self, *input, **kwargs)
    725             result = self._slow_forward(*input, **kwargs)
    726         else:
--> 727             result = self.forward(*input, **kwargs)
    728         for hook in itertools.chain(
    729                 _global_forward_hooks.values(),

~\anaconda3\envs\measurement_py37\lib\site-packages\torchvision\models\detection\generalized_rcnn.py in forward(self, images, targets)
     92                     raise ValueError("All bounding boxes should have positive height and width."
     93                                      " Found invalid box {} for target at index {}."
---> 94                                      .format(degen_bb, target_idx))
     95 
     96         features = self.backbone(images.tensors)

ValueError: All bounding boxes should have positive height and width. Found invalid box [790.0323486328125, 359.0328369140625, 790.0323486328125, 359.0328369140625] for target at index 0.
`

Answer 1

TLDR；您必须首先检查您的基本事实，并确保任何零面积框都将被丢弃

假设您提供的边界框是零面积框。 考虑到数据格式是[x1，y1，x2，y2]，它实际上表示地面真值框的[left，top…]和[…，right，bottom]边缘，在您的例子中，这两个点实际上是重叠的

这是一个有效输入的好例子，因为x1<；x2和y1<；所有盒子中的y2

result: {'boxes': tensor(*[[  0.,   0., 286., 403.]*,
         [ 30., 240.,  52., 241.],
         [ 25., 183.,  31., 204.],
         [ 26., 224.,  34., 240.],
         [ 30., 169.,  88., 181.],
         [ 32., 239.,  85., 251.],

但是错误消息说：

[790.0323486328125, 359.0328369140625, 790.0323486328125, 359.0328369140625]

这表示x1=x2和y1=y2