search

researchmm / Stark

[ICCV'21] Learning Spatio-Temporal Transformer for Visual Tracking
MIT License
649 stars 143 forks source link

Are all the models here working on single object tracking not MOT? #33

Open ayanasser opened 3 years ago

ayanasser commented 3 years ago

Hello, Thank you for your contribution ^^, I need to know if all the pretrained models here are working on single object tracking, So I wounder if there's any model working as MOT

MasterBin-IIAU commented 3 years ago

Hi, for now the STARK code only supports SOT. But it should not be too hard to extend it to MOT :)

ayanasser commented 3 years ago

Thank you @MasterBin-IIAU ^^ I have a question regarding the model demo I made, Why the tracking box is not vanishing when the tracked object is become outside the frame? Is there any parameter related to this ?

MasterBin-IIAU commented 3 years ago

@ayanasser Hi, STARK can predict a confidence score besides the bounding box. You can set a threshold such as 0.5 to control the existence of the box. For example, you can set that "if the confidence score is less than 0.5, then do not show the box"

TsingWei commented 2 years ago

@ayanasser Hi, STARK can predict a confidence score besides the bounding box. You can set a threshold such as 0.5 to control the existence of the box. For example, you can set that "if the confidence score is less than 0.5, then do not show the box"

Does STARK_X_trt predict with score besides? It seems that the output of STARK_X_trt only contains the coordinate of the box.