Please provide complete information as applicable to your setup.
• Hardware Platform: Jetson AGX Orin)
• DeepStream Version: 6.1
• JetPack Version: 5.0.2.
• TensorRT Version: 8.4.1
• Issue Type: Question
I am trying to use the image save feature.
I am calling nvds_obj_enc_process like this:
nvds_obj_enc_process(*img_save_instance->obj_enc_ctx, &userData, ip_surf, obj_meta, frame_meta);
I am getting a segmentation fault in this function call and it never returns.
Following the samples, this is my initialization of userData:
NvDsObjEncUsrArgs userData = { 0 };
/* To be set by user */
userData.saveImg = save_img;
userData.attachUsrMeta = attach_user_meta;
/* Set if Image scaling Required */
userData.scaleImg = FALSE;
userData.scaledWidth = 0;
userData.scaledHeight = 0;
/* Preset */
userData.objNum = num_rects;
/* Quality */
userData.quality = 80;
In this case, userData.fileNameImg will be initialized to 0.
If I do this instead: userData.fileNameImg[0] = '\0'
,
This function now returns, with this error: Error: Scaled object dimensions cannot be non-positive. Object not encoded.
The seg faut now occurs on the call to nvds_obj_enc_finish:
nvds_obj_enc_finish(*img_save_instance->obj_enc_ctx);
I don’t have the source code to nvds_obj_enc_process so I cannot figure out where this error is coming from. This image save feature was working as expected in Deepstream 5. Any help is appreciated, thanks.