VvasInferResult - VvasInferResult - 6.3 English - XD056

Vitis Video Analytics SDK (VVAS) 6.3 User Guide (XD056)

Document_ID
XD056
Release_Date
2026-09-24
Version
6.3 English

VvasInferResult is a type-erased container that holds the actual inference result along with utility function pointers for lifecycle management. The data pointer is cast to the concrete result structure indicated by infer_result_type.

Definition

struct VvasInferResult {
  void *data;
  VvasInferResultType infer_result_type;
  VvasInferResult* (*create) (void);
  bool (*free) (void *);
  bool (*copy) (const void *, void **);
  bool (*transform) (void *, VvasInferScaleInfo *);
};

Members

data

Pointer to the concrete result structure. Cast to VvasInferDetection when infer_result_type is VVAS_INFER_RESULT_DETECTION, or to VvasList of VvasInferClassification when it is VVAS_INFER_RESULT_CLASSIFICATION.

infer_result_type

Discriminator identifying the result type. See VvasInferResultType.

create

Allocates a new result of this type.

free

Releases the result and its data.

copy

Deep-copies the result into a new allocation.

transform

Scales bounding-box coordinates using the supplied VvasInferScaleInfo. Used internally during cascade inference to map predictions back to the original frame resolution.