search

YueLiao / PPDM

Code for "PPDM: Parallel Point Detection and Matching for Real-time Human-Object Interaction Detection".
MIT License
219 stars 42 forks source link

how to get a visualized demo (image or video) #28

Closed BestSongEver closed 4 years ago

BestSongEver commented 4 years ago

Thanx for ur great job! @YueLiao I am very interested in this work. I wonder if u can tell me how to get a visualized image demo (e.g., image from HICO, VCOCO, or own image) or video demo? I would appreciate it if you can provide related scripts.

Best regards

YueLiao commented 4 years ago

Thx for your interest in our work, and sorry for the latter response. You can take the below scripts as references. And I am too busy now to format these codes, thus the provided codes may be a bit raw.

BestSongEver commented 4 years ago

Thanx for your reply. U are such a kind researcher. @YueLiao I will try these scripts latter.

LSC333 commented 3 years ago

Thank you for these codes in visualized demo! But I noticed a line of code in the vis_predictions.py is prediction = json.load(open('/home/SENSETIME/liaoyue/ly/HOIW/vIs/model_115.json','r')). I wonder if you can tell me how to generate the json file?

YueLiao commented 3 years ago

Thank you for these codes in visualized demo! But I noticed a line of code in the _vispredictions.py is prediction = json.load(open('/home/SENSETIME/liaoyue/ly/HOIW/vIs/model_115.json','r')). I wonder if you can tell me how to generate the json file?

The 'json' can be generated by the test.py.

LSC333 commented 3 years ago

Thank you for these codes in visualized demo! But I noticed a line of code in the _vispredictions.py is prediction = json.load(open('/home/SENSETIME/liaoyue/ly/HOIW/vIs/model_115.json','r')). I wonder if you can tell me how to generate the json file?

The 'json' can be generated by the test.py.

thx! The problem is solved!

wolfworld6 commented 2 years ago

Thank you for these codes in visualized demo! But I noticed a line of code in the _vispredictions.py is prediction = json.load(open('/home/SENSETIME/liaoyue/ly/HOIW/vIs/model_115.json','r')). I wonder if you can tell me how to generate the json file?

The 'json' can be generated by the test.py.

hello,thx for the code,i run the code ,but the output_i has no information about the hoi_preds['rel_point'],why?could tell me?

YueLiao commented 2 years ago

Thank you for these codes in visualized demo! But I noticed a line of code in the _vispredictions.py is prediction = json.load(open('/home/SENSETIME/liaoyue/ly/HOIW/vIs/model_115.json','r')). I wonder if you can tell me how to generate the json file?

The 'json' can be generated by the test.py.

hello,thx for the code,i run the code ,but the output_i has no information about the hoi_preds['rel_point'],why?could tell me?

The 'rel_point' could be obtained by modifying the decode file through outputting and transforming 'ys_rel, xs_rel'.