Add --line-spacing and reduce its default

README.md

@@ -382,7 +382,7 @@ THA: Thai
                                              image folder if using batch mode
 -o, --dest DEST                              Path to the destination folder for translated images in
                                              batch mode
--l, --target-lang {CHS,CHT,CSY,NLD,ENG,FRA,DEU,HUN,ITA,JPN,KOR,PLK,PTB,ROM,RUS,ESP,TRK,UKR,VIN,ARA,CNR,SRP,HRV,THA}
+-l, --target-lang {CHS,CHT,CSY,NLD,ENG,FRA,DEU,HUN,ITA,JPN,KOR,PLK,PTB,ROM,RUS,ESP,TRK,UKR,VIN,ARA,CNR,SRP,HRV,THA,IND}
                                              Destination language
 -v, --verbose                                Print debug info and save intermediate images in result
                                              folder
@@ -448,6 +448,8 @@ THA: Thai
                                              model. Use hex string without the "#" such as FFFFFF
                                              for a white foreground or FFFFFF:000000 to also have a
                                              black background around the text.
+--line-spacing LINE_SPACING                  Line spacing is font_size * this value. Default is 0.01
+                                             for horizontal text and 0.2 for vertical.
 --force-horizontal                           Force text to be rendered horizontally
 --force-vertical                             Force text to be rendered vertically
 --align-left                                 Align rendered text left
@@ -461,7 +463,7 @@ THA: Thai
                                              additional typesetting. Ignores some other argument
                                              options
 --gpt-config GPT_CONFIG                      Path to GPT config file, more info in README
---mtpe                                       Turn on/off machine translation post editing (MTPE) on
+--use-mtpe                                   Turn on/off machine translation post editing (MTPE) on
                                              the command line (works only on linux right now)
 --save-text                                  Save extracted text and translations into a text file.
 --save-text-file SAVE_TEXT_FILE              Like --save-text but with a specified file path.

README_CN.md

@@ -114,7 +114,7 @@ THA: Thai
                                              image folder if using batch mode
 -o, --dest DEST                              Path to the destination folder for translated images in
                                              batch mode
--l, --target-lang {CHS,CHT,CSY,NLD,ENG,FRA,DEU,HUN,ITA,JPN,KOR,PLK,PTB,ROM,RUS,ESP,TRK,UKR,VIN,ARA,CNR,SRP,HRV,THA}
+-l, --target-lang {CHS,CHT,CSY,NLD,ENG,FRA,DEU,HUN,ITA,JPN,KOR,PLK,PTB,ROM,RUS,ESP,TRK,UKR,VIN,ARA,CNR,SRP,HRV,THA,IND}
                                              Destination language
 -v, --verbose                                Print debug info and save intermediate images in result
                                              folder
@@ -180,6 +180,8 @@ THA: Thai
                                              model. Use hex string without the "#" such as FFFFFF
                                              for a white foreground or FFFFFF:000000 to also have a
                                              black background around the text.
+--line-spacing LINE_SPACING                  Line spacing is font_size * this value. Default is 0.01
+                                             for horizontal text and 0.2 for vertical.
 --force-horizontal                           Force text to be rendered horizontally
 --force-vertical                             Force text to be rendered vertically
 --align-left                                 Align rendered text left
@@ -193,7 +195,7 @@ THA: Thai
                                              additional typesetting. Ignores some other argument
                                              options
 --gpt-config GPT_CONFIG                      Path to GPT config file, more info in README
---mtpe                                       Turn on/off machine translation post editing (MTPE) on
+--use-mtpe                                   Turn on/off machine translation post editing (MTPE) on
                                              the command line (works only on linux right now)
 --save-text                                  Save extracted text and translations into a text file.
 --save-text-file SAVE_TEXT_FILE              Like --save-text but with a specified file path.

manga_translator/args.py

@@ -134,6 +134,7 @@ def _format_action_invocation(self, action: argparse.Action) -> str:
 parser.add_argument('--font-size-offset', default=0, type=int, help='Offset font size by a given amount, positive number increase font size and vice versa')
 parser.add_argument('--font-size-minimum', default=-1, type=int, help='Minimum output font size. Default is image_sides_sum/200')
 parser.add_argument('--font-color', default=None, type=str, help='Overwrite the text fg/bg color detected by the OCR model. Use hex string without the "#" such as FFFFFF for a white foreground or FFFFFF:000000 to also have a black background around the text.')
+parser.add_argument('--line-spacing', default=None, type=int, help='Line spacing is font_size * this value. Default is 0.01 for horizontal text and 0.2 for vertical.')
 
 g = parser.add_mutually_exclusive_group()
 g.add_argument('--force-horizontal', action='store_true', help='Force text to be rendered horizontally')

manga_translator/manga_translator.py

@@ -542,11 +542,11 @@ async def _run_text_rendering(self, ctx: Context):
         # manga2eng currently only supports horizontal left to right rendering
         elif ctx.renderer == 'manga2eng' and ctx.text_regions and LANGAUGE_ORIENTATION_PRESETS.get(
                 ctx.text_regions[0].target_lang) == 'h':
-            output = await dispatch_eng_render(ctx.img_inpainted, ctx.img_rgb, ctx.text_regions, ctx.font_path)
+            output = await dispatch_eng_render(ctx.img_inpainted, ctx.img_rgb, ctx.text_regions, ctx.font_path, ctx.line_spacing)
         else:
             output = await dispatch_rendering(ctx.img_inpainted, ctx.text_regions, ctx.font_path, ctx.font_size,
                                               ctx.font_size_offset,
-                                              ctx.font_size_minimum, not ctx.no_hyphenation, ctx.render_mask)
+                                              ctx.font_size_minimum, not ctx.no_hyphenation, ctx.render_mask, ctx.line_spacing)
         return output
 
     def _result_path(self, path: str) -> str:

manga_translator/rendering/__init__.py

@@ -98,6 +98,7 @@ async def dispatch(
     font_size_minimum: int = 0,
     hyphenate: bool = True,
     render_mask: np.ndarray = None,
+    line_spacing: int = None
     ) -> np.ndarray:
 
     text_render.set_font(font_path)
@@ -113,14 +114,15 @@ async def dispatch(
         if render_mask is not None:
             # set render_mask to 1 for the region that is inside dst_points
             cv2.fillConvexPoly(render_mask, dst_points.astype(np.int32), 1)
-        img = render(img, region, dst_points, hyphenate)
+        img = render(img, region, dst_points, hyphenate, line_spacing)
     return img
 
 def render(
     img,
     region: TextBlock,
     dst_points,
     hyphenate,
+    line_spacing,
 ):
     fg, bg = region.get_font_colors()
     fg, bg = fg_bg_compare(fg, bg)
@@ -142,6 +144,7 @@ def render(
             bg,
             region.target_lang,
             hyphenate,
+            line_spacing,
         )
     else:
         temp_box = text_render.put_text_vertical(
@@ -151,6 +154,7 @@ def render(
             region.alignment,
             fg,
             bg,
+            line_spacing,
         )
     h, w, _ = temp_box.shape
     r_temp = w / h
@@ -177,12 +181,12 @@ def render(
     img[y:y+h, x:x+w] = np.clip((img[y:y+h, x:x+w].astype(np.float32) * (1 - mask_region) + canvas_region.astype(np.float32) * mask_region), 0, 255).astype(np.uint8)
     return img
 
-async def dispatch_eng_render(img_canvas: np.ndarray, original_img: np.ndarray, text_regions: List[TextBlock], font_path: str = '') -> np.ndarray:
+async def dispatch_eng_render(img_canvas: np.ndarray, original_img: np.ndarray, text_regions: List[TextBlock], font_path: str = '', line_spacing: int = 0) -> np.ndarray:
     if len(text_regions) == 0:
         return img_canvas
 
     if not font_path:
         font_path = os.path.join(BASE_PATH, 'fonts/comic shanns 2.ttf')
     text_render.set_font(font_path)
 
-    return render_textblock_list_eng(img_canvas, text_regions, size_tol=1.2, original_img=original_img, downscale_constraint=0.8)
+    return render_textblock_list_eng(img_canvas, text_regions, line_spacing=line_spacing, size_tol=1.2, original_img=original_img, downscale_constraint=0.8)

manga_translator/rendering/text_render.py

@@ -297,10 +297,10 @@ def put_char_vertical(font_size: int, cdpt: str, pen_l: Tuple[int, int], canvas_
         canvas_border[pen_border[1]:pen_border[1]+bitmap_b.rows, pen_border[0]:pen_border[0]+bitmap_b.width] = cv2.add(canvas_border[pen_border[1]:pen_border[1]+bitmap_b.rows, pen_border[0]:pen_border[0]+bitmap_b.width], bitmap_border)
     return char_offset_y
 
-def put_text_vertical(font_size: int, text: str, h: int, alignment: str, fg: Tuple[int, int, int], bg: Optional[Tuple[int, int, int]]):
+def put_text_vertical(font_size: int, text: str, h: int, alignment: str, fg: Tuple[int, int, int], bg: Optional[Tuple[int, int, int]], line_spacing: int):
     text = compact_special_symbols(text)
     bg_size = int(max(font_size * 0.07, 1)) if bg is not None else 0
-    spacing_x = int(font_size * 0.2)
+    spacing_x = int(font_size * (line_spacing or 0.2))
 
     # make large canvas
     num_char_y = h // font_size
@@ -659,10 +659,10 @@ def put_char_horizontal(font_size: int, cdpt: str, pen_l: Tuple[int, int], canva
 
 def put_text_horizontal(font_size: int, text: str, width: int, height: int, alignment: str,
                         reversed_direction: bool, fg: Tuple[int, int, int], bg: Tuple[int, int, int],
-                        lang: str = 'en_US', hyphenate: bool = True):
+                        lang: str = 'en_US', hyphenate: bool = True, line_spacing: int = 0):
     text = compact_special_symbols(text)
     bg_size = int(max(font_size * 0.07, 1)) if bg is not None else 0
-    spacing_y = int(font_size * 0.2)
+    spacing_y = int(font_size * (line_spacing or 0.01))
 
     # calc
     # print(width)

manga_translator/rendering/text_render_eng.py

@@ -52,12 +52,13 @@ def render_lines(
     canvas_w: int,
     font_size: int,
     stroke_width: int,
+    line_spacing: int = 0.01,
     fg: Tuple[int] = (0, 0, 0),
     bg: Tuple[int] = (255, 255, 255)) -> Image.Image:
 
     # bg_size = int(max(font_size * 0.1, 1)) if bg is not None else 0
     bg_size = stroke_width
-    spacing_y = int(font_size * 0.01)
+    spacing_y = int(font_size * (line_spacing or 0.01))
 
     # make large canvas
     canvas_w = max([l.length for l in textlines]) + (font_size + bg_size) * 2
@@ -338,8 +339,7 @@ def render_textblock_list_eng(
     font_color = (0, 0, 0),
     stroke_color = (255, 255, 255),
     delimiter: str = ' ',
-    align_center=True,
-    line_spacing: float = 1.0,
+    line_spacing: int = 0.01,
     stroke_width: float = 0.1,
     size_tol: float = 1.0,
     ballonarea_thresh: float = 2,
@@ -499,7 +499,7 @@ def update_enlarged_xyxy(region):
             line.pos_x -= canvas_x1
             line.pos_y -= canvas_y1
 
-        textlines_image = render_lines(textlines, canvas_h, canvas_w, font_size, sw, font_color, stroke_color)
+        textlines_image = render_lines(textlines, canvas_h, canvas_w, font_size, sw, line_spacing, font_color, stroke_color)
         rel_cx = ((canvas_x1 + canvas_x2) / 2 - rx) / resize_ratio
         rel_cy = ((canvas_y1 + canvas_y2) / 2 - ry + y_offset) / resize_ratio