Comparing Pre-trained Models (available on github) of Object Tracking and Re-Identification Project

I am looking for an experienced freelancer to compare different pre-trained models for object tracking, and to carry out quantitative analysis and data visualization. The main objective is to evaluate the performance of several pre-trained models, including SMILEtrack, BoT-SORT, Deep-oc-sort, Trackformer, and FairMOT, on custom inputs and the MOT20 dataset, and analyze the results using MOTChallenge metrics

You will be responsible for:

*Loading and evaluating the pre-trained models on custom inputs and a subset of the MOT20 dataset
*Implementing the pre-trained models for object tracking and re-identification
*Generating the results of the evaluation and analyzing the performance using MOTChallenge metrics
*Reporting on each method's results, including a full code, preferably using Google Colab.
Note that we do not require running the full MOT20 dataset, and a subset of the dataset is sufficient as long as the results are reliable.

links to models
A. https://github.com/WWangYuHsiang/SMILEtrack
B. https://github.com/NirAharon/BoT-SORT
C. Deep-oc-sort- https://github.com/gerardmaggiolino/deep-oc-sort
D. https://github.com/timmeinhardt/trackformer
E. fairmot -any models from github