Results 19 comments of Ayush Baid

This seems to do better ![ba](https://github.com/borglab/gtsfm/assets/3755532/9267c4cb-3383-4f09-ba03-1413234a95b5) ![ra](https://github.com/borglab/gtsfm/assets/3755532/0d604f7a-c417-44a7-8fcd-e5dc557c17c8)

> @ayushbaid the diff is against "MST w/o Huber", not "no MST" right? Are we still seeing regressions when compared to "no MST"? The metrics are compared against master though

> Why did you turn off inlier weighting? It was just for experimentation and comparing different sets of values. Inlier weighting is on right now on master.

I cannot make sense of the results. For datasets where GTSFM performs poorly, we do good. However, we see some regression on other datasets. ![rotation_avergaging_result](https://github.com/borglab/gtsfm/assets/3755532/97307b97-81e0-4ff6-8f7b-61b32f14f388) ![ba_result](https://github.com/borglab/gtsfm/assets/3755532/39ce360a-304e-4769-b346-3bef62c82a39)

Metrics on 01/02: [visual_comparison_dashboard.zip](https://github.com/borglab/gtsfm/files/13811971/visual_comparison_dashboard.zip)

> It looks like you deleted a couple of the front-end configs. Did you mean to do that? I thought everything used unified, but apparently not. Let me revert.

This is the top level python script: https://github.com/zhangganlin/GlobalSfMpy/blob/main/scripts/sfm_pipeline.py

@stepanyanhayk is there any performance implication from this?

@johnwlambert this should be closed now?

Do we still want to merge this?