Yolov8 medium model trained on about 3k Manga, Webtoon, Manhua and Western Comic style(very few) Images for text detection and segmentation. Training Imgsize = 1024. Training Images were resized, not cropped. It can handle the extreme Image ratios often found in Korean webtoons.
- Downloads last month
- 0
Unable to determine this model's library. Check the
docs
.