Travis Driver

> > > could you post a link to the colab? > > > > > > The link is at the top of the notebook > > cannot see...

> Do we still want to merge this? I think having the option of using the robust optimization is useful. Even though it does not currently show any improvements, it...

Very promising, but still probably needs some tuning (especially for SIFT). There's a lot to learn from this paper.  [visual_comparison_dashboard.zip](https://github.com/borglab/gtsfm/files/8199552/visual_comparison_dashboard.zip)

Considering >2 views for outlier rejection purposes is a great idea. Would it be possible to make this work for _N_ views? Why stop at 3?

@johnwlambert Thank you for taking a closer look at AstroNet. I think there is clearly some degradation in the results by the looks of the plots (would probably be more...

Ok, but `two_view_report` and `two_view_report_pp` are now just exact copies of each other (with `two_view_report_pp` being `None` if there are not enough inliers). I'm not sure that just copying a...

I agree that `v_corr_idxs` and `reproj_error_gt_model` should probably be removed. I'm ambivalent about removing `v_corr_idxs_inlier_mask_gt` because it's used to [visualize the metrics](https://github.com/borglab/gtsfm/blob/a7462f07a77250da0f0e913a16d23ea053ed800d/gtsfm/utils/viz.py#L264).

Thank you guys for the feedback. @akshay-krishnan Sorry, I think my description was a little misleading. The main change I made was having the loader create an instance of `GtsfmData`...

@johnwlambert I believe this has been resolved?

@johnwlambert Can we close this?