ODVTrack

The main difference between our tracker and existing trackers as follows:

The main structure of the tracker is as follows:

The speed and parameter information of the tracker are as follows:

Install the environment

Just like the environment of OSTrack, installation can be done following the method described here.

Data Preparation

Put the tracking datasets in ./data. It should look like this:

${PROJECT_ROOT}
 -- data
     -- lasot
         |-- airplane
         |-- basketball
         |-- bear
         ...
     -- got10k
         |-- test
         |-- train
         |-- val
     -- coco
         |-- annotations
         |-- images
     -- trackingnet
         |-- TRAIN_0
         |-- TRAIN_1
         ...
         |-- TRAIN_11
         |-- TEST

Set project paths

Run the following command to set paths for this project

python tracking/create_default_local_file.py --workspace_dir . --data_dir ./data --save_dir .

After running this command, you can also modify paths by editing these two files

lib/train/admin/local.py  # paths about training
lib/test/evaluation/local.py  # paths about testing

Train ODVTrack

torchrun --nproc_per_node 8 lib/train/run_training.py --script odvtrack --config crossvitb_256_ep300_template3_time --save_dir .

Evaluation

Set the trained weights path at params.checkpoint in lib/test/parameter/odvtrack.py

Change the corresponding values of lib/test/evaluation/local.py to the actual benchmark saving paths

Some testing examples:

• LaSOT or other off-line evaluated benchmarks (modify --dataset correspondingly)

python tracking/test.py odvtrack crossvitb_256_ep300_template3_time --dataset lasot --threads 16 --num_gpus 4
python tracking/analysis_results.py # need to modify tracker configs and names

• GOT10K-test

python tracking/test.py odvtrack crossvitb_256_ep300_template3_time --dataset got10k_test --threads 16 --num_gpus 4

• TrackingNet

python tracking/test.py odvtrack crossvitb_256_ep300_template3_time --dataset trackingnet --threads 16 --num_gpus 4