Skip to main content
. Author manuscript; available in PMC: 2024 Aug 22.
Published in final edited form as: Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit. 2023 Aug 22;2023:11525–11535. doi: 10.1109/cvpr52729.2023.01109

Figure 2.

Figure 2.

The flows of the two groups of latent features (categorical and directional) in the latent space of DconnNet, are visualized by T-SNE [13]. They were first disentangled (Sec 3.2) and then effectively fused in a projected shared manifold (Sec 3.3). The colors are rendered based on the results of clustering.