This commit is contained in:
2025-06-23 11:27:52 +08:00
commit b7956bd8bc
4 changed files with 26 additions and 0 deletions

3
.gitmodules vendored Normal file
View File

@ -0,0 +1,3 @@
[submodule "repo/Awesome-Optical-Flow"]
path = repo/Awesome-Optical-Flow
url = https://github.com/hzwer/Awesome-Optical-Flow

22
README.md Normal file
View File

@ -0,0 +1,22 @@
# Pose Estimation
## Optical Papers
See [Multiple-Camera 3D Motion Tracking with Optical Flow](docs/Multiple-Camera%203D%20Motion%20Tracking%20with%20Optical%20Flow.pdf) first,
to understand what I want to achieve.
See [hzwer/Awesome-Optical-Flow](repo/hzwer/Awesome-Optical-Flow/README.md) for more papers.
Following optical flow papers are good starting points.
### Supervised Models
- [FlowFormer++: Masked Cost Volume Autoencoding for Pretraining Optical Flow Estimation](https://github.com/XiaoyuShi97/FlowFormerPlusPlus)
- [FlowFormer: A Transformer Architecture for Optical Flow](https://github.com/drinkingcoder/FlowFormer-Official/)
- [Global Motion Aggregation](https://github.com/zacjiang/GMA)
- [Recurrent All Pairs Field Transforms for Optical Flow](https://github.com/princeton-vl/RAFT)
### Multi-Frame Supervised Models
- [Dense Optical Tracking: Connecting the Dots](https://github.com/16lemoing/dot)
- [VideoFlow: Exploiting Temporal Cues for Multi-frame Optical Flow Estimation](https://github.com/XiaoyuShi97/VideoFlow)