Junwei Liang

Are the videos put under "forking_paths_dataset/multifuture_visualization/"?

@MZ82020 The warnings are normal. That is due to the video encoding missing some frames at the end of the video. And I follow my instructions to generate the visualization...

tmot algorithm itself is very lightweight and the object detection and tracking is very fast. The multi-threaded code in [here](https://github.com/JunweiLiang/Object_Detection_Tracking/blob/master/obj_detect_tracking_multi_queuer_tmot.py#L733) reads the frames using CPU meanwhile running object detection on...

Please refer to the code here: https://github.com/JunweiLiang/Object_Detection_Tracking/blob/master/multi_video_reid.py It is missing some steps to produce necessary files and I currently don't have time to add them. Please only use it as...

Thanks. I don't plan on releasing the full code and instructions for this part in the near future. Please check back later.

Thanks! I will take a look & test it by the end of next week.

Sorry, been busy with intern stuff. Give me three days so I could make sure the Yii and python updates don't break anything. Also, could you put the docker stuff...

If you don't already know, [this paper](https://arxiv.org/abs/2111.02114) uses the released model to inversely get the closest 400M image-text pairs from web crawl dataset.

@AaronYALai @interxuxing I think it should the next parents should be gather_helper(input_t.parents, parents) as the parent_idx is the traceback to the last timestep's parent at each timestep.

I have downloaded the features but they seem to be a single file. Are they a single pickle binary with dictionary keys? How to read them and map them to...