Athul Suresh comments

Results 10 comments of


Athul Suresh

Can we directly implement Posefix on Inference phase or should we have to do it from training itself?

what I understood is posefix is used to reduce the error in converting heatmaps to keypoints. So is it possible to apply posefix on heatmap generated by any pose estimation...

Is it possible to use deep sort algo for tracking instead of using your simple tracker can you provide me some info regarding that?

Have you tried this with deepsort @Anupama-898 ?

Label error displayed . Which person dose the lable 'P276' indicate ?

I tried this on a surveillance cam, the id of people get swapped when two people cross path. Also if in frame t 2 person is detected and frame t+1...

Model trained on walking/loitering data

does it work on multiperson?

Annotate Our Custom Data set to Training Model

Have you figured it out? Any annotation tools for the purpose

if d not in matched_indices[:, 0]: TypeError: tuple indices must be integers or slices, not tuple

In sort.py use "from sklearn.utils.linear_assignment_ import linear_assignment" itself if you are using spacy alternate for linear assignment and downgrade the sklearn to 0.21 or lower (a version lower than 0.23)

No Difference between yolov4-tiny.cfg(csp) in this repo and darknet original yolov4-tiny.cfg

yolov4_tiny in darknet repo and in this repo is exactly the same, can i get yolov4-tiny-csp elsewhere? @WongKinYiu

About multi people to classify

have u done LSTM on multi-skeletal datapoints? @hwlee96 @mengzhangjian

P4 80x40 10S (1:10 scan) multiplexer help

> > same here, qiangli p4 80x40 rgb matrix. i'm using "passive-3" adapter and has 4 blank lines at middle (after 16 rows) and at bottom (after 36 rows) any...

how to skip detection for every n (eg. n=3) frames while keep tracking for every frame?

have u figured it out ?