LVOS

Long-term Video Object Segmentation

Download LVOS V2 Benchmark

For each video in LVOS V2, we provide images, pixel-wise annotations, and attribute annotations. Images and annotations are at 6fps .

Training Set

Google Drive | Baidu Pan

You can use the training set to train your own long-term VOS model. And we believe that the complex motion in LVOS is also helpful for short-term VOS. You can also try to train short-term VOS model on LVOS training set.

Evaulation Set

Google Drive | Baidu Pan

All the images and groundtruth masks are provided. Please use our evaluation tookit to assess your own algorithms.

Testing Set

Google Drive | Baidu Pan

Note: only images and the groundtruth masks are provided in testing set. The rest groundtruch masks are kept private. For V2 version, we do not have the plan to publish the evaluation server currently.

Meta Jsons

Download the meta jsons from the following links:

Google Drive | Baidu Pan

Download the jsons with attributes from the following links:

Google Drive | Baidu Pan

Download LVOS V1 Benchmark

For each video in LVOS V1, we provide images, pixel-wise annotations, and lingual description. Images and annotations are at 6fps .

Training Set

Google Drive | Kaggle | Baidu Pan

You can use the training set to train your own long-term VOS model. And we believe that the complex motion in LVOS is also helpful for short-term VOS. You can also try to train short-term VOS model on LVOS training set.

Evaulation Set

Google Drive | Kaggle | Baidu Pan

All the images and groundtruth masks are provided. Please use our evaluation tookit to assess your own algorithms.

Testing Set

Google Drive | Kaggle | Baidu Pan

Note: only images and the groundtruth masks are provided in testing set. The rest groundtruch masks are kept private. Please use the evaluation server for convenience of evaluating your own algorithms.

Meta Jsons

Download the meta jsons from the following links:

Google Drive | Kaggle | Baidu Pan

Download the expression jsons from the following links:

Google Drive | Kaggle | Baidu Pan

Evaulation Metrics and Toolkit

In LVOS, we utilize three metrcs: Region Similarity, Contour Accuracy, and Standard Deviation to measure the performance of VOS methods. The definitions of the three metrics can be seen in paper.

The evaluation toolkit for evaluation set can be found on Github. Please see the repository for more information.

The groundtruth masks of testing set are keep private. Please use the evaluation server instead.

We provide the tools and test scripts in this repository. Click on this link for more information.

Contact Us

Any questions, suggestions and feedback are welcomed. Please concat honglyhly@gmail.com