yangyangyang127 / PointCLIP_V2

[ICCV 2023] PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning
MIT License
222 stars 15 forks source link

How do you generate the Figure 3 in the paper? #18

Open auniquesun opened 11 months ago

auniquesun commented 11 months ago

In the paper, Figure 3 provides intuitive differences between PointCLIP V1 and V2.

I wonder how do you generate and visualize this figure. In my impression, the paired depth map and text description are both vectors, how do you use them to make this figure?

Thanks.

Harr7y commented 6 months ago

Waiting for a reply...