search

myfavouritekk / T-CNN

ImageNet 2015 Object Detection from Video (VID)
MIT License
370 stars 152 forks source link

Need window_size_7_time_step_1_top_ratio_0.000300_top_bonus_0.400000_optflow_max_pooling to receate the demo #2

Closed jeremysxy closed 8 years ago

jeremysxy commented 8 years ago

It seems there is no window_size_7_time_step_1_top_ratio_0.000300_top_bonus_0.400000_optflow_max_pooling to recreate the demo

Do I miss something?

myfavouritekk commented 8 years ago

@jeremythu The window_size_7_time_step_1_top_ratio_0.000300_top_bonus_0.400000_optflow_max_pooling directory is generated after the spatial max-pooling in the demo. Maybe something wrong with the tracking? Have you successfully compiled the tracker?

jeremysxy commented 8 years ago

This directory is ok~ I thought it's an intermediate training results or something......you can close this issue now ~ thanks a lot~