Trance-0/NoteNextra-origin
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

updates

Browse Source
This commit is contained in:
Trance-0
2025-09-18 12:44:53 -05:00
parent f62e2bb8e1
commit a6e92115f7
6 changed files with 369 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
content/CSE5519/CSE5519_G2.md
View File

@@ -2,4 +2,20 @@
## Pixel-Perfect Structure-from-Motion with Featuremetric Refinement
[link to the paper](https://arxiv.org/pdf/2108.08291)
[link to the paper](https://arxiv.org/pdf/2108.08291)
Leveraging dense local information to refine sparse observations. It is inherently amenable to SfM as it can optimize all locations over multiple views in a track simultaneously.
Both **bundle** and **keypoint adjustments** are based on geometric observations, namely keypoint locations and flow, but do not account for their respective uncertainties.
Learned representation: SfM can handle image collections with unconstrained viewing conditions exhibiting large changes in terms of illumination, resolution, or camera models. The image representation used should be robust to such changes and ensure an accurate refinement in any condition. We thus turn to features computed by deep CNNs, which can exhibit high invariance by capturing a large context, yet retain fine local details.
> [!TIP]
>
> This paper is a good example of how to use deep features for SfM with CNN and do the bundle adjustment and keypoint adjustment over the predicted features for better results.
>
> It seems to be techniques behind the scene of the first topic that interests me when I joined the computer vision class. The collection of cameras and predicted cloud points really impressed me.
>
> With RANSAC and subpixel estimation we have pretty decent results for 3D reconstruction that is scalable with noise detection.
>
> I'm a bit curious about the performance of the model in more complicated scenes like structures of a tree or other natural scenes. How the model deal with high frequency details if we fit the "smooth surface" too much?

6
content/CSE5519/CSE5519_H2.md
View File

@@ -3,3 +3,9 @@
## Does Object Recognition Work for Everyone?
[link to the paper](https://arxiv.org/pdf/1906.02659)
> [!TIP]
>
> This paper is a good example revealing the data bias problem in computer vision. I was a bit shocked by the fact that the data is not proportional to the real world human distribution.
>
> I'm a bit curious about is there any "fair" dataset that is available to the general public today? What are the metrics used to evaluate the fairness of the dataset?