quarterMoon

论文里提到的互信息损失，似乎在开源代码里没有提到。

Uncaught exception:check failed output.m() == output_width() while press the button of "mesh it"

1

PS D:\intern\kanjing\new\instant-ngp> ./buildxtt/testbed --scene data/nerf/fox 15:00:51 INFO Loading NeRF dataset from 15:00:51 INFO data\nerf\fox\transforms.json 15:00:51 SUCCESS Loaded 50 images of size 1080x1920 after 0s 15:00:51 INFO cam_aabb=[min=[1.0229,-1.33309,-0.378748], max=[2.46175,1.00721,1.41295]] 15:00:51 INFO...

About obtaining uvo data set ap or boundary ap

5

I would like to ask if the uvo data set is a video segmentation data set. How is instance segmentation performed? There are two types of videos in the uvo...

About the number of detect boxes

2

I would like to ask whether the focalnet-dino and vit-det detectors used in the paper only select the first 100 detection boxes？Thank you.

The ap of lvis measured by the public model is inconsistent with the ap in the paper.

1

When testing the ap of the lvis data set, whether to combine the output of hqsam and the output mask of sam itself to test the ap. Is sam's prediction...

Why are the indicators of inference that I use the published sam-hq-l model different from those shown in the paper? The IOU of the published model inference is 89.22. The...