Commit
•
747a7cb
1
Parent(s):
1730015
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,81 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
tags:
|
| 4 |
+
- object-detection
|
| 5 |
+
- object-tracking
|
| 6 |
+
- video
|
| 7 |
+
- video-object-segmentation
|
| 8 |
+
inference: false
|
| 9 |
---
|
| 10 |
+
|
| 11 |
+
# unicorn_track_tiny_mask
|
| 12 |
+
|
| 13 |
+
## Table of Contents
|
| 14 |
+
- [unicorn_track_tiny_mask](#-model_id--defaultmymodelname-true)
|
| 15 |
+
- [Table of Contents](#table-of-contents)
|
| 16 |
+
- [Model Details](#model-details)
|
| 17 |
+
- [How to Get Started with the Model](#how-to-get-started-with-the-model)
|
| 18 |
+
- [Uses](#uses)
|
| 19 |
+
- [Direct Use](#direct-use)
|
| 20 |
+
- [Downstream Use](#downstream-use)
|
| 21 |
+
- [Misuse and Out-of-scope Use](#misuse-and-out-of-scope-use)
|
| 22 |
+
- [Limitations and Biases](#limitations-and-biases)
|
| 23 |
+
- [Training](#training)
|
| 24 |
+
- [Training Data](#training-data)
|
| 25 |
+
- [Training Procedure](#training-procedure)
|
| 26 |
+
- [Evaluation Results](#evaluation-results)
|
| 27 |
+
- [Environmental Impact](#environmental-impact)
|
| 28 |
+
- [Citation Information](#citation-information)
|
| 29 |
+
|
| 30 |
+
|
| 31 |
+
<model_details>
|
| 32 |
+
|
| 33 |
+
## Model Details
|
| 34 |
+
|
| 35 |
+
Unicorn accomplishes the great unification of the network architecture and the learning paradigm for four tracking tasks. Unicorn puts forwards new state-of-the-art performance on many challenging tracking benchmarks using the same model parameters. This model has an input size of 800x1280.
|
| 36 |
+
|
| 37 |
+
- License: This model is licensed under the apache-2.0 license
|
| 38 |
+
- Resources for more information:
|
| 39 |
+
- [Research Paper](https://arxiv.org/abs/2111.12085)
|
| 40 |
+
- [GitHub Repo](https://github.com/MasterBin-IIAU/Unicorn)
|
| 41 |
+
|
| 42 |
+
</model_details>
|
| 43 |
+
|
| 44 |
+
<uses>
|
| 45 |
+
|
| 46 |
+
## Uses
|
| 47 |
+
|
| 48 |
+
#### Direct Use
|
| 49 |
+
|
| 50 |
+
This model can be used for:
|
| 51 |
+
|
| 52 |
+
* Single Object Tracking (SOT)
|
| 53 |
+
* Multiple Object Tracking (MOT)
|
| 54 |
+
* Video Object Segmentation (VOS)
|
| 55 |
+
* Multi-Object Tracking and Segmentation (MOTS)
|
| 56 |
+
|
| 57 |
+
<Eval_Results>
|
| 58 |
+
|
| 59 |
+
## Evaluation Results
|
| 60 |
+
|
| 61 |
+
LaSOT AUC (%): 67.7
|
| 62 |
+
BDD100K mMOTA (%): 39.9
|
| 63 |
+
DAVIS17 J&F (%): 68.0
|
| 64 |
+
BDD100K MOTS mMOTSA (%): 29.7
|
| 65 |
+
|
| 66 |
+
|
| 67 |
+
</Eval_Results>
|
| 68 |
+
|
| 69 |
+
<Cite>
|
| 70 |
+
|
| 71 |
+
## Citation Information
|
| 72 |
+
|
| 73 |
+
```bibtex
|
| 74 |
+
@inproceedings{unicorn,
|
| 75 |
+
title={Towards Grand Unification of Object Tracking},
|
| 76 |
+
author={Yan, Bin and Jiang, Yi and Sun, Peize and Wang, Dong and Yuan, Zehuan and Luo, Ping and Lu, Huchuan},
|
| 77 |
+
booktitle={ECCV},
|
| 78 |
+
year={2022}
|
| 79 |
+
}
|
| 80 |
+
```
|
| 81 |
+
</Cite>
|