jwlim / tracker_benchmark Goto Github PK

I don't know where is main_*function when I look at the following readme in https://github.com/jwlim/tracker_benchmark/tree/pami15
Could you tell me please? *main_SRE.m is for the SRE test
*main_TRE.m is for the TRE test
*main_OPE.m is for the OPE test
OPE is the first trial of TRE

setup_seqs(loadSeqs)

NameError: name 'setup_seqs' is not defined

no such file in the scripts/butil directory !!!!!!!!!

i copy the images "car4" to "data/car4" directory and then change tb_50.txt files to this :

Car4    IV ,SV

then try to run this code but i get these errors :

PS C:\Users\Mohammad\Desktop\tracker_benchmark-master> python run_trackers.py -s tb50
  Setup sequences ...
    Car4    ./data/Car4/img/
Input Test name : Mohammad
  Starting benchmark for 24 trackers, evalTypes : ['OPE', 'SRE', 'TRE']
     1_ASLA, 1_Car4:1/1 - OPE
    Starting matlab engine...
       Undefined function or variable 'vl_setup'.

Error in run_ASLA (line 5)
vl_setup;

      failed to execute ASLA : (<class 'matlab.engine.MatlabExecutionError'>, MatlabExecutionError("\n  File C:\\Users\\Mohammad\\Desktop\\tracker_benchmark-
          master\\trackers\\ASLA\\run_ASLA.m, line 5, in run_ASLA\nUndefined function or variable            'vl_setup'.\n",), <traceback object at 0x00000000036E6548>)
          Traceback (most recent call last):
       File "run_trackers.py", line 146, in <module>
           main(sys.argv[1:])
          File "run_trackers.py", line 50, in main
trackers, seqs, evalType, shiftTypeSet)
         File "run_trackers.py", line 138, in run_trackers
            butil.save_seq_result(seqResults)
         File "C:\Users\Mohammad\Desktop\tracker_benchmark-master\scripts\butil\load_results.py", line 5, in save_seq_result
          tracker = result[0].tracker
        IndexError: list index out of range

Hi,
Could you please add a README/tutorial that explains how to use the code?
Functions are documented, but it's not clear what to do first and which is the entry point.

An example tutorial that explains step-by-step what to do would be greatly appreciated.

Hi,
I run the run_trackers.py, it occurs ModuleNotFoundError: No module named 'score', I don't know how to fix it, could you tell me what should i do?

could you offer the code of Precision plot?

Hi, I tried to reproduce results from several papers.
I took their raw results and wrote my code so that this repo can evaluate performance from the raw results.
I visualized the results in terms of the bounding boxes positions in each frames.
Everything seems to work fine.
However, when I plot the graphs, it seems that the reported scores are different compared to those reported in the papers by ~1-2%.

Does anyone has the same observation?

This needs a sorted() around it to function properly. Otherwise the file names come in out of order.

when "json.dumps" result rarise error under the envrionment windows python2.7. and print message:
" [<type 'str'>, <type 'unicode'>, <type 'int'>, <type 'int'>, <type 'str'>, <type 'long'>, <type 'NoneType'>, <type 'str'>, <class 'matlab.mlarray.double'>, <type 'list'>] "

Hi! Are the results of existing trackers available in JSON format? Is there a script to convert the .mat files to .json files? Thank you!

Finally I have modified the code from python2.7 to python3.6, and solved the problem of running ASLS. BUT when running the BSBT, realized that the code was written in Windows system. It would be nice if the author could mention it in the readme.md file.

When I ran this demo, I ran 'python run_tracker.py' and then there are some wrong, could you give me some suggestion?
failed to execute LOT : (<type 'exceptions.TypeError'>, TypeError('unsupported data type unicode',), <traceback object at 0x7fd4c7c87f38>) Traceback (most recent call last): File "run_trackers.py", line 146, in <module> main(sys.argv[1:]) File "run_trackers.py", line 50, in main trackers, seqs, evalType, shiftTypeSet) File "run_trackers.py", line 138, in run_trackers butil.save_seq_result(seqResults) File "/home/hanzhj/ZTL_projects/Tracking/tracker_benchmark/scripts/butil/load_results.py", line 5, in save_seq_result tracker = result[0].tracker IndexError: list index out of range

I can draw the overlap threshold gragh? But do not know the location error threshold.Could you do me a favor ?

Hi! I noticed that the links on the website are e.g. http://cvlab.hanyang.ac.kr/tracker_benchmark/seq/Basketball.zip
while the Python toolkit downloads e.g. http://cvlab.hanyang.ac.kr/tracker_benchmark/seq_new/Basketball.zip

Is there some difference between seq and seq_new?

the first line in tb_50.txt is "BasketBall" , while the dir is "Basketball"

Hi, thank you for your source code! And I want to evaluate the algorithm on OTB-100, and there is a imitOmit.txt files included in the tracker_benchmark_v1.0, and where can I find the imitOmit files of the OTB100 dataset?

In matlab/RunTrackers.m, the init_rect is the same for all sub-sequences in TRE mode.

Perhaps this could be fixed by adding

test_cfgs(i).init_rect = seq.gt_rect(test_cfgs(i).img_start, :);

after

test_cfgs(i).img_range_str = ...

1. in butil/load_results.py line63, there is an undefined class Attribute, should it be class Score?
2. in butil/eval_results.py line84, there is an undefined variable precisionList, should it add after line55 like this: precisionList = []?
   Thank you for your reply as you are the benchmark providers!!

load_results.py ,tracker = result[0].tracker,IndexError: list index out of range.

Hello, thank you for your contributions 👍

• However, I find some differs between the version of matlab and this on attributions
• for examples:
  • the sequence of Coke in matlab, has 5 attributions which represent as [1,1,0,1,0,0,1,1,0,0,0],
    while in python version, it has 6 attributions represented as this
• some other sequences has the same problems 😢
• so, which is right? 😕

Looking forward to your reply , thanks

Best regards

directories Football1, Crossing and Couple in the data subdir are not empty

评估SRE和TRE只能在tb50上吗？为什么没有tb100的。我看TRE/SRE文件夹里面每个算法文件夹里面只有tb50的结果。

Using single tracker for -t parameter caused index out of range error in python

Hi, @jwlim I tested the code and found the requirement as stated above .Should there be at least two trackers for comparation as designed or could it be fixed?

And is there an effective way to view ground truth along with tracking result bounding box at the same time without coding in this benchmark?

That's another question, thx for sharing the code.