Skip to content

Commit f8eafbf

Browse files
committed
Merge branch 'master' of https://github.com/penghouwen/TracKit
2 parents 716bef4 + af8530b commit f8eafbf

File tree

1 file changed

+9
-2
lines changed

1 file changed

+9
-2
lines changed

README.md

Lines changed: 9 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,7 @@
88
## News
99
:boom: Implementation of our tracking and segmentation work, based on **Pytorch** and **TensorRT**.
1010

11-
:boom: The initial version is released, including [Ocean]() and [SiamDW(CVPR2019)](http://openaccess.thecvf.com/content_CVPR_2019/papers/Zhang_Deeper_and_Wider_Siamese_Networks_for_Real-Time_Visual_Tracking_CVPR_2019_paper.pdf).
11+
:boom: The initial version is released, including [Ocean](https://arxiv.org/abs/2006.10721) and [SiamDW(CVPR2019)](http://openaccess.thecvf.com/content_CVPR_2019/papers/Zhang_Deeper_and_Wider_Siamese_Networks_for_Real-Time_Visual_Tracking_CVPR_2019_paper.pdf).
1212

1313
:boom: We provide a TensorRT implementation, running at 1.5~2.5 times faster than pytorch version (e.g. 149fps/68fps for video `twinnings`, see [details](https://github.com/researchmm/TracKit/tree/master/lib/tutorial/)).
1414

@@ -43,7 +43,7 @@
4343

4444

4545
### Ocean
46-
**[[Paper]]() [[Raw Results]](https://drive.google.com/file/d/1vDp4MIkWzLVOhZ-Yt2Zdq8Z_Z0rz6y0R/view?usp=sharing) [[Training and Testing]](https://github.com/researchmm/TracKit/tree/master/lib/tutorial/Ocean) [[Demo]](https://www.youtube.com/watch?v=83-XCEsQ1Kg&feature=youtu.be)** <br/>
46+
**[[Paper]](https://arxiv.org/abs/2006.10721) [[Raw Results]](https://drive.google.com/file/d/1vDp4MIkWzLVOhZ-Yt2Zdq8Z_Z0rz6y0R/view?usp=sharing) [[Training and Testing]](https://github.com/researchmm/TracKit/tree/master/lib/tutorial/Ocean) [[Demo]](https://www.youtube.com/watch?v=83-XCEsQ1Kg&feature=youtu.be)** <br/>
4747
Official implementation of the Ocean tracker. Ocean proposes a general anchor-free based tracking framework. It includes a pixel-based anchor-free regression network to solve the weak rectification problem of RPN, and an object-aware classification network to learn robust target-related representation. Moreover, we introduce an effective multi-scale feature combination module to replace heavy result fusion mechanism in recent Siamese trackers. This work also serves as the baseline model of OceanPlus. An additional **TensorRT** toy demo is provided in this repo.
4848
<img src="https://github.com/penghouwen/TracKit/blob/master/demo/Ocean_overview.jpg" height="300" alt="Ocean"/><br/>
4949

@@ -94,6 +94,13 @@ booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVP
9494
month = {June},
9595
year = {2019}
9696
}
97+
98+
@article{zhangOcean,
99+
title={Ocean: Object-aware Anchor-free Tracking},
100+
author={Zhipeng Zhang, Houwen Peng, Jianlong Fu, Bing Li, Weiming Hu},
101+
journal={arXiv preprint arXiv:2006.10721
102+
year={2020}
103+
}
97104
```
98105

99106
## References

0 commit comments

Comments
 (0)