Increasing error after running bundle adjustment demo

Question

Increasing error after running bundle adjustment demo

lishiyu005 opened this issue 8 months ago · 4 comments

Dear author,

I try to run the script suo_slam/tree/main/thirdparty/g2opy/python/examples/object_slam_demo.py. However, I find the pose error even increase after bundle adjustment. Is there something wrong?

Thank you very much.

Best regards

Answer 1 · 2024-05-16T20:53:58.000Z

Hello,

I Don't think I dove too deeply into that script and only used it for debugging the object pose factors. We found however that the BA helps when running the actual system. Sorry I cannot provide any further reasoning.

Answer 2 · 2024-05-16T21:15:44.000Z

I also read the paper, but do not find the evaluation of multiple view. Only in table 1, but the results of multiple-view object poses are not better than GDRnet. Did I get it wrong?

Answer 3 · 2024-05-16T22:05:34.000Z

We evaluated multiview on each dataset. We are competitive GDRNet when is trains a different net for each object, but we beat it when using a single net for all objects like we do which is shown in table 1.

Answer 4 · 2024-05-17T00:29:35.000Z

Also note that we did not use the pbr data that GDRNet used, which could further boost the performance.