search

minghanqin / LangSplat

Official implementation of the paper "LangSplat: 3D Language Gaussian Splatting" [CVPR2024 Highlight]
https://langsplat.github.io/
Other
573 stars 61 forks source link

The configuration of eval code for differenct scenes #43

Open ChengZY opened 2 months ago

ChengZY commented 2 months ago

I tested on 3D-OVS data. It can achieve similar results (IoU) on the sofa scene.
But the performance for other scenes are not good as expeted.

I noted the language feature images are well-trained, the reason should be the setting of the eval code, such as threshold and the kernal size. Does it mean we need to try the setting manually to achieve the best results?

Below is the sample of the bench scene, including language feature image, groundtruth and the predicted mask. Do you have any suggestion?

sample

ChengZY commented 1 month ago

The predicted language feature image is the 512 dim language picture. To visualize the result, I selected 3 channels as RGB.