rgba images : resized alpha channel fix (graphdeco-inria/gaussian-splatting#1193)

Author: syt06162

utils/general_utils.py

@@ -19,6 +19,25 @@ def inverse_sigmoid(x):
     return torch.log(x/(1-x))
 
 def PILtoTorch(pil_image, resolution):
+    # Adapted from: https://github.com/graphdeco-inria/gaussian-splatting/pull/1193
+    # Original author: @ndming
+    # License: For non-commercial, research and evaluation use (see LICENSE.md)
+
+    # When resizing RGBA, PIL pre-multiplies the resulting RGB with the resized alpha channel. This gives
+    # different training behaviors depending on whether the image is actually resized (via -r flag) or not.
+    # Moreover, the resized alpha is no longer a perfect binary image due to interpolation, which produces
+    # a significant amount of floaters along the edges. To fix this, we manually mask the RGB if the input
+    # is an RGBA, then we forget the alpha channel entirely. The multiplication of the rendered image with
+    # the alpha_mask during training thus becomes a no-op for RGBA.
+    if pil_image.mode == 'RGBA':
+        from PIL import Image
+        image_np = np.array(pil_image)
+        rgb_np   = image_np[..., :3]
+        alpha_np = image_np[..., 3:]
+        masked_rgb_np = (rgb_np / 255.0) * (alpha_np / 255.0)
+        masked_rgb_np = np.clip(masked_rgb_np, 0.0, 1.0)
+        pil_image = Image.fromarray((masked_rgb_np * 255).astype(np.uint8))
+
     resized_image_PIL = pil_image.resize(resolution)
     resized_image = torch.from_numpy(np.array(resized_image_PIL)) / 255.0
     if len(resized_image.shape) == 3: