Update README.md

README.md

@@ -29,16 +29,6 @@ output.zip/
 
 ---
 
-## Preprocessing Pipeline
-
-- The original dataset contained approximately 23,000 face images.
-- A total of 12,000 images were selected **sequentially** (not randomly) to ensure determinism and avoid duplicates.
-- Each image underwent the following processing steps:
-  - Face region was detected and cropped using a custom ONNX-based face detection model.
-  - Images were resized to 128×128 pixels using bilinear interpolation.
-  - Each output image was renamed in sequential order as `1.jpg`, `2.jpg`, ..., `12000.jpg`.
-
----
 
 ## Usage Example