f_direction: h x w x 2 pos_list: [[y, x], [y, x], [y, x] ...] binary_tcl_map: h x w

Start recognition threads.

如果字符串尾部重复出现某个元素超过threshold次，则只保留keep个该元素 参数: s: 输入字符串 threshold: 重复次数阈值，默认20 keep: 保留的重复次数，默认5 返回: 处理后的

update

polygons (numpy.array, required): of shape (num_instances, num_points, 2)

convert text-index into text-label.

PPHGNet_base Args: pretrained: bool=False or str. If `True` load pretrained parameters, `False` otherwise. If str

PPHGNet_small Args: pretrained: bool=False or str. If `True` load pretrained parameters, `False` otherwise. If st

PPHGNet_tiny Args: pretrained: bool=False or str. If `True` load pretrained parameters, `False` otherwise. If str

Constructs a MobileNetV3-Large model

Constructs a MobileNetV3-Large model

Constructs a MobileNetV3-Large model

Constructs a MobileNetV3-Large model

Execute the task with appropriate parameters. For 'det' task: Args: image_path: Path to image or directo

对输入图像进行处理，并返回处理结果。 Args: img_path (str, optional): 图像文件路径。默认为 None。 img_numpy_list (list, optional): 图像数据列表，

执行版面检测 Args: images: 单张或多张图像 threshold: 置信度阈值 Returns: 检测结果列表

Unified interface for OpenDoc inference. Args: img_path: Path to input image or PDF file (str or Path) img_n

Process image for model input. Args: image: PIL Image Returns: dict with 'pixel_values' as numpy ar

Unified interface for UniRec inference. Args: img_path: Path to input image or PDF file (str or Path) img_nu

img_path: str, optional, default=None Path to the directory containing images or the image filename. save_dir: str, optio

调用函数，处理输入图像，并返回识别结果。 Args: img_path (str, optional): 图像文件的路径。默认为 None。 img_numpy_list (list, optional): 包含多个

batch: a list produced by dataloaders. image: np.ndarray of shape (N, C, H, W). ratio_list: np.ndarray of shape(N,2

batch: a list produced by dataloaders. image: np.ndarray of shape (N, C, H, W). ratio_list: np.ndarray of shape(N,2

Resize an image using OpenCV: - If the image is smaller than the target size, pad it to 640x640. - If the image is larger tha

Args: preds: Tensor (Batch, Seq_Len) or Dict batch: Raw batch data Returns: list: List of tuples