Fixed FBRS runtime dimension error on images with alpha channel (#5384)

CHANGELOG.md

@@ -82,6 +82,7 @@ non-ascii paths while adding files from "Connected file share" (issue #4428)
 - Visibility and ignored information fail to be loaded (MOT dataset format) (<https://github.com/opencv/cvat/pull/5270>)
 - Added force logout on CVAT app start if token is missing (<https://github.com/opencv/cvat/pull/5331>)
 - Missed token with using social account authentication (<https://github.com/opencv/cvat/pull/5344>)
+- Fixed FBRS serverless function runtime error on images with alpha channel (<https://github.com/opencv/cvat/pull/5384>)
 - Attaching manifest with custom name (<https://github.com/opencv/cvat/pull/5377>)
 
 ### Security

serverless/pytorch/saic-vul/fbrs/nuclio/main.py

@@ -24,7 +24,7 @@ def handler(context, event):
     neg_points = data["neg_points"]
     threshold = data.get("threshold", 0.5)
     buf = io.BytesIO(base64.b64decode(data["image"]))
-    image = Image.open(buf)
+    image = Image.open(buf).convert('RGB')
 
     mask, polygon = context.user_data.model.handle(image, pos_points,
         neg_points, threshold)