Due to the fast-growing applications of AI technologies applied to vehicle re-identification tasks, there have already been hot contests, such as the Nvidia AI challenge.

What algorithms or models are really adopted in commercial vehicle re-identification tasks, effective and reliable, nowadays?

  • $\begingroup$ The main problem in image recognition is to model uncertainty. A possible strategy to overcome the issue is semantic segmentation. This technique combines existing neural networks with a textual layer for scene labeling. And additional, it's important to know, that the nvidia AI contest is part of comercial efforts, because it helps to sell graphics card to a mass market. $\endgroup$ – Manuel Rodriguez Jul 10 '19 at 15:14
  • $\begingroup$ Could you please recommand some not hard to use and controllable but sufficient open source project and related papers that are most suitable to my situation, commercial vehicle re-identification? $\endgroup$ – piratesailor Jul 11 '19 at 1:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.