Motion-based post-processing: Using Kalman Filter to Exclude Similar Targets in Underwater Object Tracking

Our method

Our code will be released after the manuscript is received

Evaluation

Raw results can be found here [Baidu Drive] (password: 0000) [Google Drive]

Download UOT100(https://www.kaggle.com/datasets/landrykezebou/uot100-underwater-object-tracking-dataset)

Download UTB180(https://www.kaggle.com/datasets/bastech/utb180)

Put UOT100 and UTB in ./data. It should look like:

${PROJECT_ROOT}
 -- data
     -- UOT100
         |-- AntiguaTurtle
         |-- ArmyDiver1
         |-- ArmyDiver2
         ...
     -- UTB180
         |-- Video_0001
         |-- Video_01
         |-- Video_0002
         ...

Merge the provided code with original [OSTrack] framework

Go to lib/test/evaluation/local.py to set datasets dir

Then you can test your tracker in UOT100 and UTB180 or evaluate our raw results

• UOT100

Put the UOT100 raw results on $PROJECT_ROOT$/output/test/tracking_results/

python tracking/analysis_results.py # need to modify tracker configs and names

• UTB180

Put the UTB180 raw results on $PROJECT_ROOT$/output/test/tracking_results/

python tracking/analysis_results.py # need to modify tracker configs and names

Acknowledgments

• Thanks for the OStrack library, which helps us to quickly implement our ideas.