Documentation improvements.

Emgu.CV.Contrib/DepthAI/Device.cs

@@ -22,8 +22,11 @@ namespace Emgu.CV.Dai
     public partial class Device : UnmanagedObject
     {
         /// <summary>
-        /// Create a new device
+        /// Initializes a new instance of the <see cref="Device"/> class using the specified DepthAI pipeline.
         /// </summary>
+        /// <param name="pipeline">
+        /// The <see cref="Pipeline"/> object that defines the configuration and operations for the DepthAI device.
+        /// </param>
         public Device(Pipeline pipeline)
         {
             _ptr = DaiInvoke.daiDeviceCreate(pipeline);

Emgu.CV/Core/Mat.cs

@@ -1295,6 +1295,7 @@ namespace Emgu.CV
         /// Returns a memory span that wraps the underlying memory buffer.
         /// </summary>
         /// <param name="size">The size of the Span. If it is less than or equals to 0, the full span is used.</param>
+        /// <typeparam name="T">The type of the elements in the span.</typeparam>
         /// <returns>A memory span that wraps the underlying memory buffer.</returns>
         public Span<T> GetSpan<T>(int size = 0) where T : struct
         {

Emgu.CV/Dnn/DnnInvoke.cs

@@ -319,10 +319,18 @@ namespace Emgu.CV.Dnn
         private static extern IntPtr cveReadNetFromTensorflow2(IntPtr bufferModel, int lenModel, IntPtr bufferConfig, int lenConfig);
 
         /// <summary>
-        /// Reads a network model stored in TFLite framework's format.
+        /// Reads a network model stored in the TensorFlow Lite (TFLite) framework's format.
         /// </summary>
-        /// <param name="model"></param>
-        /// <returns>Net object</returns>
+        /// <param name="model">
+        /// The file path to the TFLite model.
+        /// </param>
+        /// <returns>
+        /// A <see cref="Net"/> object representing the loaded network.
+        /// </returns>
+        /// <remarks>
+        /// This method loads a pre-trained neural network model stored in the TFLite format.
+        /// The model can then be used for inference or further processing.
+        /// </remarks>
         public static Net ReadNetFromTFLite(String model)
         {
             using (CvString modelStr = new CvString(model))