Changkun Liu

  Is this 1*7 vector has the same meaning as what we can get from Colmap? i.e, X Y Z = TX TY TZ, W P Q R =...

The meaning of inliers in visualization code

1

What is the difference between the two approaches, f'found {ret["num_inliers"]}/{len(ret["inliers"])} inlier correspondences.' and 'visualization.visualize_loc_from_log(images, query, log, model,top_k_db=10)', the inlier number and ratio seem cannot match?   

I can use the ` include_top` parameter in `EfficientNet.from_name`, but in `EfficientNet.from_pretrained`, I cannot use that. So, if I want to change the last layer of the pretrained model by...

Hi Siyan, Thanks for your summary work. I would suggest to add a new paper from 2023 AAAI in pose regression. https://arxiv.org/abs/2211.11238 Best, Changkun

You may can also add this paper in pose regression section

2

2022 ECCV: DFNet: Enhance Absolute Pose Regression with Direct Feature Matching https://github.com/ActiveVisionLab/DFNet

I trained with different prompt with your pretrained model. I felt that it could only generate melody, but could not generate songs with lyrics.