I followed the "sample re-identification algorithm plus the evaluation code" instructions in README.md. When running fullPipelineScript.m, MATLAB crashes almost immediately with this message:
Using default parameters
=============== Parameters for experiment v001: ================
Training Cameras: [17 18 19 40 50 53 54 56 57 58 59]
Test Cameras : [60]
Detector name : AcfInria
Use/train False Positive class? (YES/NO) : 1
Filter detections by overlap? (YES/NO) : 1
Maximum overlap beween detections threshold : 0.3
Re-Identifier name : BhattacharryaNNReId
Feature extraction name : extractHSVfromBodyParts
Using training set : /mnt/windows7/Users/vislab/Desktop/gsaponaro/HDA_Dataset_V1.1/hda_sample_train_data
================================================================
Saved allFP.txt to /mnt/windows7/Users/vislab/Desktop/gsaponaro/HDA_Dataset_V1.1/hda_experiment_data/AcfInria_001/FPClass
Saved allUniqueFP.txt to /mnt/windows7/Users/vislab/Desktop/gsaponaro/HDA_Dataset_V1.1/hda_experiment_data/AcfInria_001/FPClass
crop.m: Not creating the cropped images, letting the re-identifier crop on-the-fly
crop: Working on camera: 60
Loaded allD.txt from /mnt/windows7/Users/vislab/Desktop/gsaponaro/HDA_Dataset_V1.1/hda_detections/AcfInria/camera60/Detections
??? Index exceeds matrix dimensions.
Error in ==> seqReaderPlugin>getTs at 188
fseek(fid,info.seek(frame+1),'bof');
Error in ==> seqReaderPlugin>open at 128
ts = getTs( 0:(n-1), fid, info );
Error in ==> seqReaderPlugin at 48
[infos{h},fids(h),tNms{h}]=open(fName,info); return;
Error in ==> seqIo>reader at 114
r=@seqReaderPlugin; s=r('open',int32(-1),fName);
Error in ==> seqIo at 65
case {'reader','r'}, out = reader( fName, varargin{:} );
Error in ==> crop at 84
seqReader = seqIo( seqName, 'reader'); % Open the input image sequence
Error in ==> fullPipeLineScript at 64
crop();
I followed the "sample re-identification algorithm plus the evaluation code" instructions in README.md. When running
fullPipelineScript.m
, MATLAB crashes almost immediately with this message: