SA-Gate Segmentation - 3.5 English

Vitis AI Library User Guide (UG1354)

Document ID
Release Date
3.5 English

SA-Gate is a neural network that is used for indoor segmentation. The input is a pair of an RGB image and an HHA map generated with the depth map. The output is a heat map where each pixel is predicted with a semantic category, like chair, bed, and other objects typically found indoors.

The following image shows the result of SA-Gate segmentation.

Figure 1. SA-Gate segmentation Test Example 1

Figure 2. SA-Gate segmentation Test Example 2

Figure 3. SA-Gate segmentation Test Example 3

The following table lists the SA-Gate models supported by the Vitis AI Library.

Table 1. SA-Gate Segmentation Models
No Model Name Framework
1 SA_gate_pt PyTorch