Will the MOTA change for DLA-34 and HRNetV2-18 on same dataset ?

ifzhang / FairMOT

[IJCV-2021] FairMOT: On the Fairness of Detection and Re-Identification in Multi-Object Tracking

MIT License

4k stars 933 forks source link

Will the MOTA change for DLA-34 and HRNetV2-18 on same dataset ? #157

Closed dpsdev14091998 closed 4 years ago

ifzhang commented 4 years ago

Yes, it will change. The MOTA is affected by the training model and testing parameters.

apekshapriya commented 4 years ago

Hey @ifzhang Which one is usually better on a human-based dataset?

ifzhang commented 4 years ago

@apekshapriya I test them on the test set of MOT17 dataset and find DLA-34 has better results. The deformable convolution in DLA-34 plays an important role. HRNet-W18 has better performance on small objects.

apekshapriya commented 4 years ago

@ifzhang Thanks!