Minghan Qin

icon indicating a website link https://minghanqin.github.io

icon company Tsinghua University icon location I am interested in 3D vision, with a keen interest in areas such as 3D Scene Reconstruction and Perception and 3D Human Avatar Reconstruction. 😄

Results 6 comments of Minghan Qin

Can someone help me with the following? After I set up the environment, I cannot understand how to start quickly

Hi, @KohsukeIde This is a bug from an old version of the code; I have fixed it. Please update to the latest version. Thanks for your attention!

display result

Hi, We have open-sourced the evaluation part of the code. Our demo video comes from a modified version of the eval code. The specific modifications include: 1. Separating text queries...

The experiment reproduction question for 3D semantic segmentation on the 3D-OVS

Hi, Regarding the calculation of mIoU for the "sofa" scene, we calculate the average relevancy scores within the mask region for each relevancy map and select the relevancy map with...

The experiment reproduction question for 3D semantic segmentation on the 3D-OVS

Hi, We have open-sourced the evaluation part of the code!

About the Ground Truth

I seek clarification regarding your reference to "GT"; are you referring to the visualization result of the feature map? It is imperative to acknowledge that if the feature map's output...

Querying the most relevant 3D Gaussians

Thanks for your attention. The eval code has been released.