BNU-IVC / FastPoseGait

FastPoseGait is a user-friendly and flexible repository that aims to help researchers get started on pose-based gait recognition quickly.
64 stars 3 forks source link

Questions about the paper and implementation #6

Closed z915287285 closed 1 year ago

z915287285 commented 1 year ago

Hello, I have some questions after reading your paper and hope to get your answers. 1.Temporal correlation verification. Have you considered doing some verification of the time contribution of the TCN module? 2.About Oumvlp Dataset. The Oumvlp dataset exists in the verification part of the reference code. Are there any experimental results of Oumvlp-pose?

DreamShibei commented 1 year ago

Hi! Thanks for your interest.

  1. Considering the set-based method is more robust to the condition of discontinuous frames or different frame rates, which is more applicable in real-world scenarios, we finally choose not to use the TCN module in our work.
  2. OUMVLP-Pose is extracted from RGB frames of the OUMVLP dataset. We have reported the results of OUMVLP-Pose both in GPGait and FastPoseGait.
z915287285 commented 1 year ago

Thank you for your answer!