YOLOv8 model for detecting speech bubbles in manga pages. Trained to locate bubble bounding boxes (single class location-of-bubbles).
Part of
Manga Translate — a full manga translation pipeline (bubble detection → OCR → inpaint → LLM → render).
1from ultralytics import YOLO
2
3model = YOLO("bubbles_detect.pt")
4results = model.predict("page.jpg", conf=0.25, iou=0.5, imgsz=1024)
Or use it automatically via the
Manga Translate app — the model is downloaded from here on first run.
Trained on a combined dataset published at
PSImera/manga_bubbles_detect:
Training scripts and reports:
training/
