issues
search
axinc-ai
/
ailia-models
The collection of pre-trained, state-of-the-art AI models for ailia SDK
2k
stars
318
forks
source link
Update object_detection/layout_parsing
#1386
Closed
ohshima-axell
closed
7 months ago
ohshima-axell
commented
7 months ago
object_detection/layout_parsing の修正を行いました。変更は以下の2点です。
pdf2image の import 位置を調整し、画像を入力するだけであれば pdf2image のインストールを必須でないようにしました
infer_from_pdf() 関数内で定義されている compute() 関数に args.detector 引数の値による分岐がありますが、layout_parsing には --detector オプションがないため、infer_from_image() 関数と同様の実装に変更しました
object_detection/layout_parsing の修正を行いました。変更は以下の2点です。