inference: drop mv_facial_landmark_set_model call 04/317404/2
authorInki Dae <inki.dae@samsung.com>
Thu, 2 Jan 2025 02:04:42 +0000 (11:04 +0900)
committerInki Dae <inki.dae@samsung.com>
Mon, 13 Jan 2025 04:56:20 +0000 (13:56 +0900)
Drop mv_facial_landmark_set_model call.

mv_facial_landmark_set_model API is a private one for internal developers.
As for custom model support, we will introduce a new interface to support
custom model later.

Change-Id: Ie118aaf33b79fef49c9bfccaa556807b75457940
Signed-off-by: Inki Dae <inki.dae@samsung.com>
inference/backends/mediavision/src/MvFaceLandmark.cpp

index 74b0db44782481f5bb3065d84943c6561e96a3ce..d09c5b6460ed905b92e398e42046e6e01c1452a0 100644 (file)
@@ -46,12 +46,7 @@ MvFaceLandmark::~MvFaceLandmark()
 
 void MvFaceLandmark::configure()
 {
-       int ret = mv_facial_landmark_set_model(_handle, "fld_u2net_160x160.tflite", "fld_u2net_160x160.json",
-                                                                                  "", "FLD_U2NET");
-       if (ret != MEDIA_VISION_ERROR_NONE)
-               throw runtime_error("Fail to set face landmark model.");
-
-       ret = mv_facial_landmark_configure(_handle);
+       int ret = mv_facial_landmark_configure(_handle);
        if (ret != MEDIA_VISION_ERROR_NONE)
                throw runtime_error("Fail to configure face landmark detection.");
 }