Run the API server

convert text-index into text-label.

Initialize and get a logger by name. If the logger has not been initialized, this method will initialize the logger by adding one or two handl

Intersection.

Return the column width of text. Correct ``len(text)`` for wide East Asian and combining Unicode chars.

convert text-index into text-label.

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

Add id for gather feature, for inference.

Compute the areas of rectangles given two corners. Args: left_top (N, 2): left top corner. right_bottom (N, 2): right bottom corne

convert text-index into text-label.

convert text-index into text-label.

img_height, img_width = img.shape[0:2] left = int(np.min(points[:, 0])) right = int(np.max(points[:, 0])) top = int(np.min(points[:,

box_score_fast: use bbox mean score as the mean score

create operators based on the config Args: params(list): a dict list, used to create some operators

CTC decoder using multiple processes.

convert text-label into text-index.

convert text-index into text-label.

convert text-index into text-label.

convert text-index into text-label.

convert text-index into text-label.

locality aware nms of EAST :param polys: a N*9 numpy array. first 8 coordinates, then prob :return: boxes after nms

Convert k points to 1 polygon. Args: points (ndarray or list): A ndarray or a list of shape (2k) that indicates k points.

Calculate the intersection area between two polygon. Args: poly_det (Polygon): A polygon predicted by detector. poly_gt (Polygon)

Generate shrink_quad_along_width.

Generate shrink_quad_along_width.

Generate shrink_quad_along_width.

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

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

transform

Sends a POST request to the translator server to translate a PDF. Parameters ---------- input_pdf_path : Path Path to the PDF to

Get parameters for PaddleOCR. Parameters ---------- model_root_dir : Path Path to the PaddleOCR model ro

Initialize OCR model. Parameters ---------- model_root_dir : Path Path to the PaddleOCR model root direc

Backend function for loading models. Called in the constructor. Load the layout model, OCR model, translation model and font.

Merge translated PDF files into one file. Merged file will be stored in the temp directory as "translated.pdf". Parameters

Split text into chunks of sentences within text_limit. Parameters ---------- text: str Text to be split.

Translate text using the translation model. If the text is too long, it will be splited with the heuristic that each sentence should

Translate one page of the PDF file. There are some heuristics to clean-up the results of translation: 1. Remove newlines, tabs, b

_break_word(word : string, space_left : int) -> (string, string) Break line by unicode width instead of len(word).

_handle_long_word(chunks : [string], cur_line : [string], cur_len : int, width : int)

mkdir if not exists, ignore the exception when multiprocess mkdir together

Backend function for translating PDF files. Translation is performed in the following steps: 1. Convert the PDF file to images

Calculate the IOU between two boundaries. Args: src (list): Source boundary. target (list): Target boundary. Returns: i

box_score_slow: use polyon mean score as the mean score

_bitmap: single map with shape (1, H, W), whose values are binarized as {0, 1}

Cluster pixels in tcl_map based on quads.

CTC decoder using multiple processes.