GroundingDetection<T>

Models & Types · AiDotNet.VisionLanguage.Grounding

A region a grounding model found for a caption.

Implements: IEquatable<GroundingDetection<T>>

Constructors

Constructor Summary
GroundingDetection(T X1, T Y1, T X2, T Y2, T Score, string Phrase) A region a grounding model found for a caption.

Properties

Property Summary
T Score { get; set; } The highest token probability of the region’s query.
T X1 { get; set; } Left edge, as a fraction of the image width.
T X2 { get; set; } Right edge, as a fraction of the image width.
T Y1 { get; set; } Top edge, as a fraction of the image height.
T Y2 { get; set; } Bottom edge, as a fraction of the image height.
string Phrase { get; set; } The caption’s words whose tokens scored above the text threshold.

Methods

Method Summary
GroundingDetection<T> <Clone>$()
bool Equals(GroundingDetection<T> other)
bool Equals(object obj)
bool op_Equality(GroundingDetection<T> left, GroundingDetection<T> right)
bool op_Inequality(GroundingDetection<T> left, GroundingDetection<T> right)
int GetHashCode()
string ToString()
void Deconstruct(T X1, T Y1, T X2, T Y2, T Score, string Phrase)