Smart feather: hard inner mask + 2px soft boundary only

app.py

@@ -913,13 +913,25 @@ def _lama_inpaint_tile(img_rgb: np.ndarray, mask: np.ndarray, size: int = 512) -
         out_img_512 = np.clip(out[0].transpose(1, 2, 0) * 255.0, 0.0, 255.0).astype(np.uint8)
         out_img_orig = cv2.resize(out_img_512, (cw, ch), interpolation=cv2.INTER_CUBIC)
 
-        # FIX: Hard mask replacement - no alpha blending to prevent text ghosts.
-        # Only pixels where crop_mask==255 are replaced with LaMa output.
-        # Pixels outside the mask keep the original artwork 100% untouched.
-        hard_mask = (crop_mask > 127).astype(np.uint8)
-        hard_mask_3ch = np.stack([hard_mask, hard_mask, hard_mask], axis=-1)
-        result_crop = np.where(hard_mask_3ch, out_img_orig, crop_img)
-        img_out[y0:y1, x0:x1] = result_crop
+        # SMART FEATHER: hard mask for ALL inner text pixels (no ghost text),
+        # + thin 2px soft feather ONLY at the outer boundary (no sharp visible edges).
+        # Step 1: inner mask = pixels that were definitely text (hard replacement)
+        inner_mask = (crop_mask > 127).astype(np.uint8)
+        # Step 2: compute 2px outer boundary ring of the mask
+        boundary_kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (5, 5))
+        dilated_boundary = cv2.dilate(inner_mask, boundary_kernel, iterations=1)
+        boundary_ring = (dilated_boundary - inner_mask).astype(np.float32)  # only outer edge
+        # Step 3: soft feather weight for the boundary ring only (0→1 over 2px)
+        feather = cv2.GaussianBlur(boundary_ring, (5, 5), 1.0)
+        feather_3ch = np.stack([feather, feather, feather], axis=-1)
+        # Step 4: inner = hard LaMa, boundary = soft blend, outside = original
+        inner_3ch = np.stack([inner_mask.astype(np.float32)] * 3, axis=-1)
+        result_crop = (
+            inner_3ch * out_img_orig.astype(np.float32)
+            + feather_3ch * out_img_orig.astype(np.float32)
+            + (1.0 - inner_3ch - feather_3ch) * crop_img.astype(np.float32)
+        )
+        img_out[y0:y1, x0:x1] = np.clip(result_crop, 0, 255).astype(np.uint8)
 
     return img_out